Reform as Routine: Organizational Change in the Modern World

REFORM AS ROUTINE This page intentionally left blank REFORM AS RO U T I N E Organizational Change and Stability i n...

Author: Nils Brunsson

30 downloads 882 Views 667KB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

REFORM AS ROUTINE

This page intentionally left blank

REFORM AS RO U T I N E Organizational Change and Stability i n t h e Mo d e r n Wo r l d

N I L S B RU N S S O N

1

3

Great Clarendon Street, Oxford ox2 6dp Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With ofﬁces in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York # Nils Brunsson 2009 The moral rights of the author have been asserted Database right Oxford University Press (maker) First published 2009 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, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose the same condition on any acquirer British Library Cataloguing in Publication Data Data available Library of Congress Cataloging in Publication Data Data available Typeset by SPI Publisher Services, Pondicherry, India Printed in Great Britain on acid-free paper by MPG Biddles, King’s Lynn, Norfolk ISBN 978–0–19–829670–6 1 3 5 7 9 10 8 6 4 2

Contents Preface Acknowledgements 1. The Reforming Organization The construction of organizations

vii ix 1 1

Reforms: frequency, contents, and consequences Reforms along fundamental institutions Reform dynamics

6 10 14

Homogeneity and heterogeneity of forms

18

2. Beliefs Creating Reform: The Case of Markets and Organizations Beliefs and practice Markets and organizations Allocation of responsibility

21 21 24 31

Bases and problems of legitimacy

33

Reforms

38

3. Constructing Organizations: The Example of Public Sector Reform

43

Reforms that construct organizations Constructing identity

45 46

Constructing hierarchy Constructing rationality

50 51

Reforms in packages or in strings Degrees of organization: actors, agents, and arenas Organizatory reform as completion Incentives for constructing organizations Conclusions

54 56 63 67 70

vi

Contents

4. Politicization and ‘Company-ization’ – On Institutional Afﬁliation and Confusion in the Organizational World The political organization and the company Politicization, company-ization, and associationization

72 73 81

Processes of institutional confusion

84

Weak institutions

89

5. Organizational Reforms as Routines The supply of problems The supply of solutions The supply of forgetfulness The risks of reform Avoiding reform 6. The Standardization of Organizational Forms as a Cropping-up Process Standardization processes A cropping-up model of organizational forms

91 92 96 98 100 101 105 107 112

Model dynamics

119

Interaction between reforms and discourse Some additional questions

123 125

7. Reform and Power Reformers as agents Reform as image building 8. The Hopeful Organization The hope for the true organization Mechanisms of hope Maintaining hope References Index

127 127 134 140 141 143 152 155 167

Preface The idea of this book was born almost ten years ago when I was invited to give the Leake lectures in accounting at Oxford University and the London School of Economics. Organizational reform was the topic I chose for these lectures. At the time, I had carried out several studies of organizational reform and published some of the ﬁndings, mostly about the dynamics of reforms. But I also thought that some questions remained. They had to do with the more fundamental conditions that could explain why there is such an abundance of organizational reform in modern society. I found one such factor in the fact that more and more organizations are created as well as in the way they are constructed. This was discussed in an article co-authored by Kerstin Sahlin-Andersson and reproduced as Chapter 3 in this book. For this analysis it was possible to use secondary data in the form of an extensive literature on public sector reform that had been produced in the 1989s and 1990s. This literature provides perhaps the most extensive, empirical, and insightful analyses of reorganization that has ever been produced. In fact, the article offers a reinterpretation of this literature. Another question was how reforms could continue in spite of an extensive collective experience with implementation problems and failures. What makes people maintain their hope for implementing reforms, even the kind of reforms that intend to create situations that very seldom, if ever, have been possible to realize. This question required much more extensive empirical work. So it took some time before I could conclude my research on organizational reform. In this book, I have collected papers about reform dynamics and about fundamental conditions for reform. The purpose is to present one consistent perspective, which also provides a way of understanding reforms that is different from the way reforms are usually presented by reformers and discussed among practitioners. At the same time, this book is intended to contribute not only to the research on reorganization but also to two other strands of research – the more general research topic of the relation

viii

Preface

between institutions and organizations as well as the more speciﬁc research on the transformation and modernization of contemporary states. Nils Brunsson

Jakobsberg, June 2008

Acknowledgements Chapter 3 is reprinted from Organization Studies, Vol. 21, No. 4, 2000, 721–46. The chapter was co-authored by Kerstin Sahlin-Andersson. Some minor, technical adjustments have been made to the original text. Chapter 4 is reprinted from Management Accounting Research, Vol. 5, 1994, 323–35. Some minor adjustments have been made to the original text. Chapter 5 is reprinted from Scandinavian Journal of Management, Vol. 22, 243–52, 2006 (1989). Some minor adjustments have been made to the original text. Chapter 6 is reprinted from Scandinavian Journal of Management, Vol. 13, No. 3, 307–20, 1997. Some minor, technical adjustments have been made to the original text. Chapter 7 is a substantially revised version of chapter 12 in Brunsson, N. & Olsen, J. P. (eds): The Reforming Organization, 2nd edn. Bergen: Fagbokforlaget, 1997. Chapter 8 draws from the book Brunsson, N.: Mechanisms of Hope. Maintaining the dream of the rational organization. Malmo¨: Copenhagen Business School Press, 2006.

This page intentionally left blank

1 The Reforming Organization Modern society is highly reformistic. There are many attempts at generating agreement between the way things ought to be and the way they are – between ideals and practice. Sometimes we think that things ought to revert to the way they once were or that they should be like they are elsewhere, and we try to reform our practice accordingly. Sometimes we are inspired more by the world of ideas than we are by the world of practice, leading to higher aspirations. We try to reform our practice or that of others to conform to our ideals, without being sure that these ideals have ever been possible to realize. Large contemporary organizations, whether public or private, seem to be under almost perpetual reform – attempts at changing organizational forms. Managers try to reorganize, introduce new departments, or change the tasks of existing departments. They try to change procedures and systems for control by introducing new systems into their budgeting or accounting, for example. Or they try to change the attitudes of organization members to become more ‘customer oriented’, more ‘service oriented’, or more caring about the effectiveness and efﬁciency of the organization. But why are so many organizational reforms being launched? Who decides the content of these reforms? And what are the consequences of reforms and reforming? These are questions I discuss in this book.

TH E CONSTRUCTION OF ORGANIZATIONS We live in what has been called an organization society (Simon 1991). There is an enormous number of formal organizations, and all over the world new organizations are formed as the solution to virtually every

2

Reform as Routine

problem (Chapter 3 of this book; Drori, Meyer, and Hwang 2006). Modern individuals spend much of their time within the framework of various organizations. Almost everyone is a citizen of a state; most of us are employed in states, ﬁrms, or associations; and many of us belong to several associations: labour unions, sport associations, churches, political parties, or social movements, for instance. In addition, there are several hundreds of thousands of organizations in which other organizations, rather than individuals, are members (Ahrne and Brunsson 2008). Organizations are central in most markets. We buy products and services almost exclusively from ﬁrms and seldom from individuals; market relations are less likely to exist between individuals than they are to exist between ﬁrms, between ﬁrms and states, or between ﬁrms and individuals. There are immense inequalities between organizations and individuals: almost all organizations are richer than almost all individuals. Organizations such as states and multinational corporations are liable to control enormous resources compared to the resources of even the richest of people. The afﬂuence of organizations makes them powerful in social life.

The true organization Although organizations are manifold, rich, and powerful, no one has ever seen an organization. In order to demonstrate the existence of an organization we cannot merely point at a person or a thing. If we are to make others (and ourselves) believe that an organization exists, we must represent it in a way that conforms to the institution of Organization – to the generally shared conceptions and legal and other rules for what is an organization. If we want others to believe that we are describing an organization, we have to claim that our organization has a clear identity; organizations have clear boundaries and special characteristics. An organization should have a clear and unique purpose or task. An organization should have certain autonomy: it should be able to make decisions and undertake actions as a result of their decisions. An organization has some type of hierarchy: a decision mechanism or a management that is able to coordinate and control actions, and by which the organization’s task or purpose is translated into action. Instrumental rationality ﬁts this model

The Reforming Organization

3

of organization better than any other type of intelligence does (Brunsson 2006). Rationality is a form of processual hierarchy, a procedure by which identity is efﬁciently transformed into action. When justifying organizational actions, it is difﬁcult to represent other forms of intelligence such as rule following and imitation, because its image of autonomy and uniqueness could be jeopardized. In short, there is an overall rule for how organizations can be represented – a rule of identity, hierarchy, and rationality. All these features are constitutive for an organization; they enable us to identify an organization. If an entity seems to lack any of these aspects, it cannot be considered to be an organization. Breaking the rule for representation has serious consequences; not only is it wrong, but it also means that the organization disappears. If a group of people is represented as having no special task, no special characteristics, no clear boundaries, or no management, for instance, or if it lacks any autonomy, most people would have difﬁculty believing that such a group is an organization. Or they would argue that this cannot be a ‘true’ or ‘proper’ organization – a ‘true’ organization is an entity that is perceived to be in close conformity with the institution of Organization. The rule of identity, hierarchy, and rationality governs not only what we can say about an organization, but also what we can want it to be and become. For a person responsible for an organization, it is difﬁcult to issue goals or mission statements arguing that the organization should strive to avoid a clear identity, hierarchy, and rationality. Should one want it to differ dramatically from these characteristics, it could be interpreted as a wish to dissolve the organization.

Incentives for reform The representation of organizations in accordance with the rule of identity, hierarchy, and rationality seems to become more and more unrealistic – more and more difﬁcult to combine with detailed and plausible descriptions of organizational practice. As the number of organizations increases, each individual organization is exposed to numerous demands from strong, organized interests. Many of these demands are conﬂicting (Brunsson 2002). It is no longer enough for a company to be proﬁtable

4

Reform as Routine

and to provide its owners with fair revenue. There is also an expectation that it will provide employment; offer good working conditions to its employees (and, increasingly, to its subcontractors); to contribute to democracy, gender and ethnic equality, growth, innovation, and progress in general; and not to pollute the environment. In short, it must satisfy many, often-conﬂicting demands. Or, as the most recent formulation goes, it shall demonstrate ‘Corporate Social Responsibility’. It is difﬁcult to maintain autonomy under such conditions, as external forces rather than the organization itself determine what the organization shall do. If the demands are similar for many organizations, the organizations become more similar than distinct. The organizational boundaries become unclear when organizations are forced to employ experts in the various ﬁelds in which demands are being made; these experts often have more in common with experts in their ﬁeld in other organizations than they have with colleagues in their own organization. It becomes difﬁcult to coordinate the various units that work for different interests, and organizations become fragmented (Pollitt 2009). The organization is controlled as much by other organizations and other people as by its own management. The objectives become so many and conﬂicting that it is difﬁcult to apply any model of rationality to calculate the optimal action. Rather organizations follow rules set by others. It is difﬁcult, in short, for organizations to live up to the rule that governs representations and wishes. They can or must operate in other ways, and this discrepancy between rule and operations is a powerful incentive for reform. A difference between what is said and what is done deﬁnes hypocrisy, and a difference between what is wanted and what is done can be deﬁned as a problem. According to conventional wisdom, hypocrisy should be avoided and problems should be solved. One way of trying to do just that is to reform practice – to make it conform to representations and wishes. There are even more incentives for organizational reform. There is an extensive general discourse in parts of society about appropriate organizational forms, inspired by the assumptions that organizational operations can be improved, that form has a substantial impact on operations, and that forms can be changed. Furthermore, there are many management gurus, scholars, consultants, and journalists who propagate various ideas about the best forms for all or for certain groups of organizations.

The Reforming Organization

5

They launch ever-new ‘recipes’ (Røvik 2007) or ‘standards’ (Brunsson and Jacobsson 2000) for how to organize: matrix organization, process organization, management by objectives, zero-based budgeting, quality systems, balanced scorecards, or process reengineering. These standards do not break with the rules for representation; most often they explicitly refer to the content of these rules (Brunsson 2006). They merely provide more detailed guidelines. Standards are accompanied by arguments that may persuade people in organizations that these standard forms are better than the organization’s existing form, thus driving claims for reform. Or standardizers may persuade outsiders with the same effect; modern organizations are scrutinized by outsiders and compared to various ideals (Power 1997). They are expected to be able to represent their forms to an external public, and this public is likely to expect the organization to follow some popular standard for organizations in general. Although there are many competing standards for organizational form, some forms become more popular than others. For a limited period many organizational experts and managers consider a certain standard to be unquestionably correct. Such standards can be deﬁned as fashions, if, ironically, they are not perceived to be fashions by their many proponents – if they are perceived to be the best and most rational way to organize (Røvik 1998). Long-range planning was such a fashion some twenty years ago (Mintzberg 1994); management by objectives has been highly fashionable several times in the past sixty years or so (Sundstro¨m 2003), and quality systems have recently been in fashion. It is possible to distinguish three major organizational types or subinstitutions – the State, the Firm, and the Association (Poliyani 1968) – any one of which may become the most fashionable over time. During periods when the State has been popular, ﬁrms have tried to introduce structures, processes, and ideas typical of states (strategies, budgets, or allegedly democratic structures, for example). When the Firm has been popular, on the other hand, states have tried to imitate ﬁrms (Czarniawska 1985). Fashions constitute stronger rules than other standards do; they become a type of norm, albeit relatively short-lived. To some extent, they inﬂuence what can be wanted and said. During their time, they are difﬁcult to avoid, providing as they do a stronger incentive for reform than do other standards. Their content becomes the basis for reform.

6

Reform as Routine

Although contemporary organizations are under particularly strong pressure to reform, reforms seem to have been a trait of formal organizations for a long time. A study of the ﬁrst 130 years of the Swedish State Railways revealed that this organization was undergoing major reorganizations every ﬁve to ten years in its early days, and that the pace of reform has substantially increased over time (Brunsson et al. 1989).

REFORMS: FREQUENCY, CONTENTS, AND CONSEQUENCES The phenomenon that I have called ‘reforms’ has many names: reorganizations, change projects, rationalization, restructuring, and sometimes, although less often, reforms or administrative reforms. All these terms denote attempts at changing organizational forms – at attempts to re-form. That reforms are attempts means, by deﬁnition, that they do not necessarily succeed. Reform is not equivalent to change. An organization may undergo several reforms and emerge with little change. During a certain limited time, some people may merely describe the organization in a new way, with no other consequences for the organization’s activities. Organizations change in ways other than by reforms. Most organizational changes occur routinely, without drama and without reform (March 1981). Organizations change with a changing environment, changing technology, or personnel turnover, even without the intervention of top management. Reforms, on the other hand, are initiated or at least supported by management. But they are far from being the only instrument for change to which management has access. Management can try to exchange personnel; pay more attention to certain projects and neglect others; redistribute resources; or create a challenging environment by increasing the expectations of customers, the general public, or other external parties. Reforms differ from such change attempts, in that they start with explicit descriptions of the state for which one is striving. Reformers describe, at least roughly, the organization they are striving for, its structure, procedures, and ideals. Certain departments with certain tasks are to exist in the future and to work in a certain way. Their members are to execute their work and allow it to be governed by objectives or quality

The Reforming Organization

7

assessments, for example; or perhaps the personnel are to think in a certain way, are to be animated by a certain business concept or logic. By describing the state they want to attain, reformers are forced into the jurisdiction of the rules for what can be wanted and said. Therefore, reforms tend to follow rules for representations in general, rules about clarity and simplicity. It is not by chance that reformers rarely describe the goal of their reforms as increased turmoil, confusion, and obscurity; rather the purpose of the typical reform is to increase order and clarity. The new organization is to follow principles that are easy to understand: it is to have a clear division of labour and clear external and internal boundaries, simpliﬁcations are to be made, and the tasks of the organization and departments are to be well-deﬁned and straightforward. Even if there are often good reasons for an organization to operate in a way that is complex, unclear, and ambiguous to an observer and experienced managers know this, such characteristics are virtually impossible to launch as the content of reform. Organizational reforms must also comply with the rules for the institution of Organization. Because they are represented, they must comply with the rules for representing organization, and because they express a will they must comply with the rules for what we want organizations to be. Many reforms are explicit attempts to establish the ‘true’ organization. Such reforms can be directed towards developing the organization’s identity, giving it a clearer task, a clearer business concept, clearer goals, or, as has been popular recently, streamlining its activities into one clear task by ‘focusing on the core business’. Alternatively, the purpose is to improve the hierarchy – to improve and increase the efﬁciency of management control with new management and accounting systems. Or, there is an attempt to increase rationality by developing numerical objectives, evaluating results, introducing better follow-up and evaluation systems and more developed budget systems, or by introducing more rational decision-making processes. Other reforms have other purposes, but must treat the rules for the ‘true’ organization as a limitation for the content. The latter includes reforms that build primarily on institutions other than the Organization – on attempts to realize the institution of the Individual or the Market, for instance, in reforms oriented towards ‘empowerment’ or internal markets. Such reforms must comply with the rules of their institutions,

8

Reform as Routine

yet they cannot be formulated to be in major conﬂict with the institution of the Organization – often a delicate balance.

Reform dynamics Organizational reforms tend to be easy to sell. By describing current organizational practice in some detail, reformers can demonstrate that it is characterized by confusion and obscurity or that the practice indicates a weak organizational identity, weak hierarchical control, or lack of rationality. In short, the organization does not work as represented by its management or as management wants it to work. This is all argument for reform, and ideas for the changes to be instated are available both at a general level (the ‘true’ organization) and at a more detailed level (in the form of organizational standards). It is difﬁcult for an opponent to argue that the organization is in perfect harmony with the idea of the true organization. It is equally difﬁcult not to want the organization to become truer or to argue that all organizational standards created by organizational experts would be useless in the opponent’s organization. It may be possible to reject a speciﬁc standard, but more difﬁcult to reject the introduction of any and all standards. Fundamentally, reforms are stimulated by the differing ways in which representations, wishes, and practice are regulated in organizations. At the same time, these very differences produce difﬁculties in implementing reforms. Those who are expected to change their way of operating are likely to ﬁnd it difﬁcult to apply the general principles of the reform to their daily work situation, to know what the reform means, or to see that it would in any way improve their work. People are used to the organization working in conﬂict with the way it is represented. Perhaps they have previously experienced similar reforms that have had little effect on their operations. Or perhaps their old way of working has proved successful in the past. But even if reformees ﬁnd the reform unrealistic or inappropriate, they have difﬁculty arguing against it as long as it is described as a set of general principles. In principle, it is right! But the opposition often turns up during implementation.

The Reforming Organization

9

Against this background, it is not surprising that the problem most often discussed in relation to reforms is their implementation, whereas the very existence of reforms is more or less taken for granted. Reformers typically treat implementation as their main problem. The difﬁculties and failures of implementation is a common theme for analysis by evaluators and scholars, and there is a rich literature treating these problems (Winter 2003). In this book, however, I discuss some aspects of organizational reform that have received less attention. Why do reforms occur and what determines how often they occur? Why do they propagate certain ideas to the exclusion of others? What consequences other than implementation do they have? In order to understand why reforms are such common phenomena, it is necessary to understand their broader cultural and institutional background. Moreover, reforms are common because beliefs about organizations can remain relatively unaffected by organizational practice and vice versa, and disappointments in reform implementation do not necessarily lead to a loss of hope for the future realization of reforms. These conditions create reformers as well as reforms. I argue that the impact of reformers on reforms is more limited than appears from the reformers’ descriptions of reforms. The content of a reform is determined more by societal institutions and public discourse than by local ‘actors’, whether individuals or organizations. As for consequences, I argue that the difﬁculty of implementation may represent a solution rather than a problem. And even if reforms often have a limited effect on the practice that they are meant to change, they can have other signiﬁcant consequences, such as creating organizations, conﬁrming and stabilizing our conceptions of organization, and giving rise to new reforms. This book is based on numerous empirical studies of reforms in companies and in public administration. In addition, it relies upon an extensive literature on public sector reform that has evolved since the 1980s and has produced a great deal of empirical data that speak to the more general literature on reforms. The book is intended to contribute to the literature on change in general and organizational reform in particular. It is also intended as a contribution to the discussion within the so-called institutional analysis

10

Reform as Routine

of how and why certain organizational forms are adopted by organizations in a variety of ﬁelds and in geographically dispersed areas. In the next three chapters, I discuss the relationship between reforms on the one hand and fundamental societal institutions such as Organization, Market, and organizational subinstitutions on the other. Chapters 5, 6, and 7 treat the dynamics of reform processes. What drives reforms, and to what results? In the last chapter, I discuss how reformers and others can continue hoping for reforms to succeed, in spite of all the difﬁculties and failures.

REFORMS ALONG FUNDAMENTAL INSTITUTIONS An underlying theme in Chapters 2 to 4 is that our institutionalized beliefs constitute an important factor stimulating reform. These beliefs are to a large extent formed in separation from practical experience and make practice seem deﬁcient and in need of reform.

Beliefs and the practice of reforms In the next chapter, I continue the discussion about the difference between the rules for what we say and the rules for what we do, now formulated as a difference between beliefs and practice within institutions, and I investigate the effects of this difference on the frequency of reforms. I compare the institution of the Organization with the institution of the Market; these institutions are closely connected, but are based on beliefs, some of which oppose each other. Beliefs about organizations and markets seem to be strongly inﬂuenced by scientiﬁc theories about these phenomena. When these theories form the contents of reforms, they can be said to ﬁll a ‘performative’ role (Callon 1998). But the performativity of scientiﬁc theories varies, some being more inﬂuential than others. Relatively simple theories describing ideal types have had a stronger inﬂuence than have more complex theories based on extensive empirical research.

The Reforming Organization

11

Beliefs about organizations or markets do not control or reﬂect practice in any direct way. Yet, they have indirect practical effects. The differences in beliefs about markets and organizations lead to differences in allocation of responsibility. Organizations concentrate responsibility: the organization’s management is given a high degree of responsibility, thereby decreasing the responsibility of other organizational members. In markets, on the other hand, responsibility is much more evenly dispersed among all individuals and organizations acting there. Organizations and markets also differ in the way legitimacy is created: organizations are more difﬁcult to combine with modern ideas about individuality than are markets. Responsibility allocation and legitimacy problems affect the frequency of reforms. Organizations produce more reformers than markets do, this being one reason for the larger number of reforms in organizations. The differences in reform activity between organizations and markets may, in turn, reinforce their belief systems – a belief that organizations are more heterogeneous and changeful than markets are, for instance.

The construction of organizations Over the past few decades there has been a tendency for large organizations such as states and corporations to split into several smaller organizations, thus contributing to a general increase in the number of formal organizations. The purpose of creating more organizations in this way is often to create more market relations. This is the case in countries that have transformed their planned economies into market economies, countries where large state-owned industrial conglomerates have been split into smaller units. In corporations in traditional market economies, it has been popular to outsource various departments, which have then established market relations with their old companies. Many states have changed their organization in a similar way. Through the movement sometimes called ‘New public management’ (Hood 1991), states have split a coherent administration into a large number of relatively autonomous organizations. Sometimes the explicit purpose has been to create more market – to make state-run units sell to and buy from

12

Reform as Routine

each other or external parties. In other cases, market relations have merely emerged as a result of reforms. Reforms intended to establish markets or ‘true’ markets stimulate attempts at creating ‘true’ organizations, and reforms intended to create ‘true’ organizations make it tempting to try to create ‘true’ markets. In Chapter 3, ‘new public management’ reforms are used to illustrate how organizations are constructed. It is argued that these reforms were much more radical than the term ‘new public management’ indicates; they created organizations that had never before existed. Reforms have formed complete organizations from departments of states and cities. Entities that were originally incomplete organizations at best have been turned into complete organizations. This is true for professional arenas, such as universities and hospitals, as well as such administrative units as police and tax ofﬁces. These bodies have been furnished with local objectives, policies, resources, and boundaries, as well as with local responsible managers. The central state has relaxed its control over these units, while their local management has gained more power, thereby challenging professional values and reducing employees’ freedom of action more than the traditional central control did. What centrally placed politicians and administrators perceive as extensive decentralization is often seen as strong centralization by the employees. Some of the newly formed organizations have been sold to private interests, for in contrast to professional arenas or administrative units, organizations can be bought and sold. Forming organizations is similar to reforming them: in both cases, an attempt is made to create a practice in accordance with the notion of the true organization. Yet the two activities should be distinguished analytically. It is one thing to explain why reforms are being initiated in an existing organization. It is quite another to try to explain the more radical step: ceasing to conceive of a large organization such as a state as one organization and creating new organizations. In the latter case, one does not comply with the institution of the organization in reforming the state as a whole, yet complies with it when constructing the new organizations. It is far from obvious why one steps outside the institution of the organization while simultaneously entering it. Chapter 3 offers some speculations about possible causes.

The Reforming Organization

13

One effect of the creation of organizations is continued reforms with a similar content – just because (and only because) something is seen as an organization, it becomes relevant and tempting to compare its practice with the idea of the true organization and to reform in that direction. Forming is followed by reforming. The creation of the organization is followed by ‘reorganization’; when it turns out that practice does not quite coincide with the new representation of the entity as an organization, there is a strong incentive to make it do so. Reforms are stimulated, furthermore, by the new afﬁliation to the extremely general category of ‘Organization’. The construction of organizations within states and cities has involved the substitution of one category – the Organization – for many more speciﬁc categories such as university, school, hospital, or tax authority. All these units have ended up in the world of organizations – a very large world of organizational experts, organizational consultants, organizational standards, and organizational fashions – and they can begin to compare themselves with and imitate other organizations. Their construction as organizations is a strong driver for adopting the same standards as private corporations and for imitating such corporations because these have a long tradition and a remarkable ability to represent themselves as organizations (Lamoreaux 2004).

‘Company-ization and politicization’ The idea of the true organization has a powerful impact on the way new organizations are formed and how they are reformed. But organizational subinstitutions play a role in reforms as well. Once an entity is perceived as an organization, it can be deﬁned as belonging to one of these types: the State (or the Political Organization), the Firm, or the Association. These subinstitutions differ in the construction of their organizational environments and inner workings. To some extent they regulate the way actual organizations can be formed. The regulation is far from perfect, however; it is conspicuously weaker than the regulation provided by the more general institution of the organization. In Chapter 4, I argue that the existence of organizational subinstitutions creates a certain freedom for contemporary organizations: in their reforms

14

Reform as Routine

they are able to borrow relatively freely from the subinstitutions, and amalgamate them. The result in many organizations is a certain degree of institutional confusion (or ‘hybridization’ as some would perhaps call it). Organizations are ‘politicized’, ‘company-ized’, and ‘association-ized’. Firms are represented by concepts from the state and the association; states use concepts belonging to the ﬁrm or the association. The types of processes that may lead to institutional confusion are also discussed in Chapter 4.

REFORM DY NAMICS Reforms are driven not only by an organization’s relationship to a wider cultural setting with special belief systems. There are also certain characteristics of organizations that stimulate reforms, and reform processes have speciﬁc dynamics of their own. Such characteristics and dynamics are discussed in Chapters 5 to 8.

Reforms as routine In Chapter 5, I argue that organizational reforms are driven by problems to be addressed, by solutions to be applied, and by forgetfulness. The greater the supply of any of these factors, the more likely it is that reforms will occur. Without problems, reforms are difﬁcult to justify; without solutions they cannot be formulated; and without forgetfulness there is a risk that people will be discouraged by the fact that similar reforms have been tried and have failed in the past. In contemporary large organizations, problems tend to be easily found. It is not merely the deviation from a true organization that poses a problem for organizations; organizations encounter many dilemmas, such as the balance between centralization and decentralization or between differentiation and integration. And even if the present seems rosy, it is not difﬁcult to point to future problems. Furthermore, the contemporary organizational world is full of solutions in the form of standards and the form of management consultants who sell solutions

The Reforming Organization

15

to organizational problems. Those interested in selling solutions often try to supply problems as well – problems that can be solved by their solutions. Forgetfulness can be promoted by the use of consultants with limited experience of the implementation and long-term effects of reforms. Reforms are likely to emerge from such general situations. Reforms are also self-referential; they tend to cause new reforms. Thus reforms can be considered as routines: they are likely to be repeated over and over again. Paradoxically, one effective way of stopping reforms is to try to implement them and propose new reforms, which is a further reason why reforms are so common.

The ‘cropping-up’ of organizational forms In Chapter 5, I discuss temporal aspects of reform, and in Chapter 6, I introduce spatial aspects. Over the past few decades, scholars have paid a great deal of attention to the issues of homogeneity and heterogeneity among organizations. Similar organizational structures have been explained by the existence of a common institutional environment or a common general discourse about organizational forms, sometimes characterized by fashions. Differences among organizations have been attributed to different positions in a process of diffusion, to differences in the way general ideas are translated within organizations, or to differences in organizational characteristics that make some organizations more susceptible to certain fashions or standards than others are. Organizational reforms have a complicated relationship with the issues of homogeneity and heterogeneity among organizations. Although reforms are often described by reformers as attempts at ﬁnding unique solutions for the organization under reform, they actually tend to follow generally shared ideas and common standards and fashions, and are therefore a strong force for creating similarities among organizations. In spite of organizations being extremely active in reforming, however, and in spite of this homogenizing effect of reforms, we can often observe substantial differences in speciﬁc organizational forms among organizations that are difﬁcult to explain using the traditional explanatory models

16

Reform as Routine

of diffusion, translation, or variations in fundamental organizational characteristics. Chapter 6 presents a different model for explaining homogeneity and heterogeneity, drawing from the analysis of reform processes presented in earlier chapters and adding an element of randomness. This model combines space and time; it explains the spatial category of homogeneity and heterogeneity as a result of time – as a result of the relationship between the timing of general fashions and the timing of local reforms.

Reforms and power Within organizations, reforms are regularly described not as routines but as major dramatic changes. The content is described as being adapted to the needs and conditions of the speciﬁc organization. Reformers tend to be described as powerful, as the main initiators of reforms, and as the people deciding the content of those reforms. The arguments in this book, however, suggest a different picture. In Chapter 7, I draw some conclusions about the role of power in reforms, arguing that reformers have limited power in reform processes. Rather than reformers creating reforms, then, reforms create reformers. Reforms are initiated by other processes than by the intervention of reformers. The content of reforms is decided by institutionalized belief systems and by standards produced by external experts rather than by reformers. And the effects of reforms on organizational practice are regularly less extensive or deep than expected and planned. Reformers are better described as agents of other forces rather than as powerful actors. The limited power of reformers does not mean that reforms are inconsequential. The most important effects of organizational reform may occur in the environment of the reformed organizations – as when reforms improve the image of the organization among external parties. While intended to change the internal workings of organizations, reforms change their environment instead – which can be equally beneﬁcial or even more beneﬁcial to the organization. Furthermore, reforms may affect reformers rather than reformees, creating greater responsibility for the reformers, for instance, and improving reformers’ abilities to represent

The Reforming Organization

17

their organizations. That a reform fails to change organizational practice is not necessarily a sign that it has had no impact or that it is useless. Finally, reforms may lead to greater stability than change. Representing controversial changes as initiated by reforms can be a way of preventing them. And reforms may stabilize and reinforce traditional beliefs about organizations.

The hopeful organization Hope is a necessary condition for reform. Reforms are based on the hope that ideas will be transformed into practice. But reforms also increase the risk that people lose hope. When we confront our ideas with practice is precisely when we may realize that they are impractical, inappropriate, or completely unrealistic. If reforms are to continue, this outcome must be avoided. Maintaining hope for the realism and usefulness of ideals typical of organizational reform is no trivial matter. In Chapter 8, I provide some examples of ‘mechanisms of hope’ that prevent people from relinquishing their ideals – that keep them hopeful and encourage them to continue reforming even when they experience great difﬁculty in implementing their reforms. People succeed in disregarding information about practice and practical effects, they attend only to practical experiences that do not contradict their ideals, or they interpret experience in a way that does not undermine the hope that the ideas of the reform can be realized in the future. Hope is maintained by such mechanisms as an interest in talk rather than practice, a strong emphasis on the far future rather than on the present and the past, a belief that the world is under rapid change, an interest in the reform of others rather than reform of oneself, and a belief in one’s own uniqueness. Mechanisms of hope are necessary ingredients in a world in which we are expected to believe in lofty ideas, in a world in which practical experience is not allowed to play an overly important role in forming our ideas, and where we are able to try repeatedly to reform our practice to conform with these ideas, even if such reforms constantly fail. In short, hope is a fundamental, cultural factor in explaining the production and reproduction of reform.

18

Reform as Routine

Mechanisms of hope are likely to exist outside the context of organizational reforms as well, which may be true for some of the other phenomena I describe in this book. Some of the analysis may be applicable to contexts of reform that are not discussed here – for understanding the high degree of reformism in modern society in general and for understanding reforms in a more general sense, whether they concern new organizational forms or something else. In Chapter 2, I point to some similarities and differences in conditions for the reform of individuals, markets, and organizations. I invite the reader to speculate about further such similarities and differences.

HOMOGENEITY AND HETEROGENEITY OF FORMS Standards prescribing new organizational forms spread easily and rapidly over large distances in the contemporary world, resulting in similarities among organizations, even if they are geographically separated and even if they have different purposes and operations (Brunsson and Jacobsson 2000; Meyer et al. 2006). Reforms and the characteristics of reforms can be used for explaining homogeneity and heterogeneity in organizational forms. In this book, I point to several circumstances that make organizations particularly likely to adopt new standards from the outside. I indicate the signiﬁcance of the tendency towards broad generalization, of common dilemmas and difﬁculties in organizations, of decisions, and of hope. First, as argued in Chapter 3, there are more organizations today than ever before. When various entities deﬁne themselves as belonging to the category of organization, they expose themselves to a large number of standards directed at this extremely general category. Second, I argue that large organizations experience dilemmas and difﬁculties that provide continuous arguments for reform (Chapter 5), providing ample opportunities for external standards to slip into the organization. Third, I emphasize the role of decisions and underline the fact that reforms represent decided attempts at change. Decision is a fundamental aspect of organizations (March and Simon 1958; Luhmann 2000). In their

The Reforming Organization

19

decision making, organizations express how it ought to be – how the world as it is should be changed. Therefore, organizations maintain the attention of two worlds, not only the world as it is, but also the world as it ought to be. Decisions constitute a kind of talk that follows the rules for talk and representation, particularly rules for how things ought to be. Decision making renders the organization sensitive to widely shared ideals of the way things should be. When such ideals are compared to organizational practice, they inspire reforms with these ideals as their contents. Decisions to reform organizations constitute relatively quick ways to signal adaptation to standard forms that are popular in the general discourse and are difﬁcult to reject openly. One need not immediately represent a practice that conforms to the standards. Decisions are also easy to represent to outsiders who may confuse them with actual practice or practice that will occur in the near future. Reforms signal change, although change of operations is an uncertain outcome of reforms. That signal makes it more plausible for other organizations to assume that a standard form is realistic and useful, which facilitates its further adoption. Fourth, the adoption of standard forms is supported by hope. Mechanisms of hope make empirical knowledge less relevant, which helps to speed the adoption and reduce the risk that standards fall into disrepute. Hope helps standards to remain on an attractively abstract level without being compromised by the complications of the everyday world. As explained in Chapter 8, hope reduces the need for proving that standard forms are, in fact, possible and useful to implement. If one can maintain the hope that standard forms will one day be realized, even if they have never been realized before, one need not wait for knowledge about the practical effects of a standard in order to embrace it. When looking at others who have embraced a standard, hope makes it relevant to learn from their decisions – not necessarily from their practice. And implementation problems in one’s own organization do not immediately (if ever) give rise to the conclusion that the standards promoted by a reform are wrong, and therefore unworthy of recommending to others. The propensity of organizations to reform, their openness to ideals and standards about how organizations in general ought to function, and their ability to hope facilitate a widespread adoption of standard forms and make for similarities among organizations. The very deﬁnition of an entity as an organization opens it up to similarities with all other

20

Reform as Routine

units that are perceived as organizations all over the world. But there are some tendencies counteracting homogeneity. Difﬁculties of and variation in implementation help organizations to maintain their differences in practice even though they produce similar talk. Any competition among the standards propagated makes heterogeneity more likely. And, as demonstrated in Chapter 6, if standard ideas are unstable they will further heterogeneity. Knowledge about organizational reform processes is crucial for understanding how homogeneity and heterogeneity of organizational forms arise.

2 Beliefs Creating Reform: The Case of Markets and Organizations Societal institutions have been deﬁned as sets of beliefs about the world, rules about how to act, and patterns of action (Berger and Luckmann 1966). Institutionalized belief systems include speciﬁc ideas about what is true, certain norms for what is right, and a speciﬁc set of intentions that can be used when rationalizing actions. Beliefs, rules, and practice are all taken for granted by a certain society or parts thereof at a certain time, i.e. they are treated as external and objective constraints by the individual actor (Jepperson 1991). Institutionalized patterns of action and most rules regulate what we can do, institutionalized beliefs regulate what we can say. Institutionalized beliefs are concepts and assumptions about the world that most people know and presume that others know, which we expect others to refer to when they talk and write, and which therefore are useful for representing and describing our own experiences. Institutionalized beliefs form a kind of ofﬁcial truth in a society. Often, but far from always, institutionalized belief systems also regulate what we think – people come to believe in the ofﬁcial truth (Douglas 1986).

BELIEFS AND PRACTICE Berger and Luckmann (1966, Ch II:1) described the three aspects of institutions as being highly interdependent. Beliefs generate rules, and rules generate patterns of action, and these patterns in turn strengthen the beliefs and rules. This implies a notion of consistency between beliefs, rules, and practices. But such consistency does not always prevail in modern institutions.

22

Reform as Routine

Instead, we often get a difference between institutionalized beliefs and practice, between how we talk and how we act (March 1986; Brunsson 2007, Ch 7). Scholars have pointed to numerous examples. In many scientiﬁc ﬁelds there is a speciﬁc set of beliefs that can be referred to when describing scientiﬁc activity and arguing for the validity of its results, which is strongly at odds with empirical descriptions of scientiﬁc practice (Kaplan 1964; Latour 1987). In some social science settings, researchers have to refer to positivistic and deductive research ideals when describing their research, even though in practice they conduct their research in a much more inductive and less organized way; there is a sharp difference between ‘reconstructed logic’ and ‘logic-in-use’ (Kaplan 1964). States routinely represent themselves as highly sovereign entities, even in the common cases when there is no such sovereignty in practice (Krasner 1999). In organizations, we get a discrepancy and ‘decoupling’ between how organizations are represented by their leaders and others, and how they actually work (Weick 1976; Meyer and Rowan 1977): representations tend to follow institutionalized ideas as to what an organization should look like, while organizational practice often does not. There is a long tradition among students of organizations to observe a difference between ‘formal’ structure and ‘informal’ structure (Barnard 1938; Scott 1998). Similar observations have been made for members of organizations in general – there is a difference between people’s espoused theories and theories-in-use (Argyris and Scho¨n 1978). Such discrepancies arise because institutionalized beliefs and institutionalized practice are regulated differently, and because beliefs and practice have little inﬂuence on each other. Beliefs are then formed in other ways than just from experience with a certain practice. At the same time, practice is carried out fairly independently of institutionalized belief systems. There are several mechanisms by which beliefs and practice are kept apart and which hinders one to inﬂuence the other. First, in modern societies much of our belief does not come from our personal experience but as general, abstract knowledge transferred via mass-education (Meyer and Jepperson 2000). The potential of such general knowledge for inﬂuencing practice is less than its potential for inﬂuencing beliefs. Practice takes place under very varied and speciﬁc local conditions. Hence, people may ﬁnd it difﬁcult to transfer very general and abstract ideas into terms applicable to their own local and complex practice. And

Beliefs Creating Reform: The Case of Markets and Organizations

23

they may have little interest in doing so, since they do not ﬁnd general knowledge particularly relevant to their own practice. Second, even if a practice is difﬁcult to describe as in accordance with institutionalized beliefs, this does not necessarily destabilize the beliefs or make the beliefs useless. They can still be useful and stable. Under modern conditions we are often expected to have knowledge and opinions about practices of which we have no personal experience. In such situations, we are likely to relate to institutionalized beliefs whatever their relation to practice. And even if we do have an actual personal knowledge of practice that deviates from institutionalized truths, this does not necessarily lead to rejection of the institutionalized truth. As observers and commentators of life in general, we need general knowledge; as practitioners we need speciﬁc knowledge. So institutionalized beliefs are useful for describing the general case, even if they cannot be used for practical purposes. Moreover, the speciﬁc knowledge we use as actors does not need to inﬂuence how we talk about the world in general. When relating our speciﬁc personal experience to the general case, we can, in principle, lean towards either of two opposite strategies. One is generalization, whereby our beliefs about what is general are formed on the basis of our own experience; the other could be called localization, whereby our own experience is explained as a special case and an exception to what is generally true. The modern situation strongly favours localization over generalization: science explicitly warns against generalizing from one or a few cases. And a well-established general truth makes it both difﬁcult and unnecessary to use the generalization strategy – most people already know what is generally true. If people see their own practice and experience as speciﬁc and exceptional, they are far less likely to disbelieve general belief systems on a basis of their own experience (see Chapter 8). The localization strategy also makes it less likely that some people’s speciﬁc knowledge will affect other people’s belief in the general knowledge: there doesn’t seem much point in reporting what is perceived as special cases to others at all, and the others may not in fact be very interested in hearing about them. And accounting for local experience that deviates from general beliefs may be unwise. The local practice might easily appear deviant, strange, or incorrect, or the accounts might sound incredible or at least difﬁcult to understand. It is often

24

Reform as Routine

safer to describe and defend local practice in terms of the general belief system, even if one thinks that it is a very poor description of the practice. So we should not be astonished when ﬁnding that institutionalized beliefs and practices sometimes do not reﬂect each other. However, this does not mean that beliefs and practices are unrelated. In this chapter, I argue that even when beliefs are poor descriptions of practice, they can still have important, although more indirect, practical effects. One such effect is the production of reforms. I illustrate this thesis by referring to institutionalized beliefs about the organization comparing it with the institutions of the market and the individual. I point to three effects of institutional beliefs: their effect on responsibility allocation, on legitimacy, and on reform. Differences between beliefs about organizations and beliefs about markets lead to differences in the way responsibility is allocated within these institutions. Institutionalized beliefs affect legitimacy and legitimacy problems of markets and organizations. Responsibility allocation and legitimacy in turn have practical effects, including an effect on reform: they inﬂuence the extent to which organizations and markets are subject to attempts at reform, as well as determining who the reformers are and the content of their reforms. In order to be able to discuss these arguments in detail, I have ﬁrst to describe some beliefs and practices in markets and organizations. This is the topic of the next section.

MARKETS AND ORGANIZATIONS

Belief systems There is a strong scientiﬁc tradition to describe both markets and organizations in very general terms as ideal types. A standard description of the market as an ideal type based on neoclassical economic theory posits two types of actor – sellers and buyers (Samuelsson 1964). These actors are given; they seem to appear without anyone’s intervention, complete with their speciﬁc interests and action, and their appearance does not demand special explanations.

Beliefs Creating Reform: The Case of Markets and Organizations

25

The seller and buyer are described as actors whose boundaries are clearly deﬁned; they possess internal consistency and speciﬁc preferences, and they act in accordance with these preferences. The boundaries must be clear if exchange is to take place; we must know when the commodity is inside or outside each actor. The exchange is driven by preferences and purposes. The sellers try to increase their proﬁts, and the buyers try to increase their utility. These are their only fundamental interests; they are therefore perceived as striving for their ‘self-interest’. This leads to a conspicuous lack of loyalty; the actors are ready to dump each other as soon as they ﬁnd they can satisfy their interests better with a new partner. On a ‘free’ but not ‘perfect’ market this is both possible and usual, since there are many possible sellers and buyers of a certain product who should work under competitive conditions. Volatility as to who are the actors is an important aspect of markets. Not only the market actors but markets themselves are taken for granted even in some more sophisticated discussions; markets appear as something natural rather than planned and created – as a ‘spontaneous order’ (Hayek 1988). Such ideas about naturalness have strong roots in older academic treatises now deemed classic: Smith (1776, Book 1, Ch 2) saw the tendency for business and bargaining as a trait of human nature and the resulting order as shaped by an invisible hand; Mandeville (1724) used the metaphor of a bee colony for describing market behaviour. In this perspective any attempts at active regulation of markets can be perceived as interferences with a ‘free’ market. The idea of naturalness of the free market signals a certain homogeneity and stability: if markets are natural we can expect their basic functioning to be more homogeneous and stable than if we assumed that they were created by different people with different interests and changed according to these people’s varying and potentially unstable ideas about how they should work. The organization, too, has been described as an ideal type (Weber 1924 (1947) Ch. III; Etzioni 1964). In these descriptions the organization is very different from the ideal description of the market. Organizations are described as rational and as instruments – their purpose is to achieve a certain common result. The organization has a task, a mission, which is either formulated on a basis of common interests among its members, or it is decided by its principals. Either way, the organization does not arise naturally, it has to be created by someone with a purpose in mind. The

26

Reform as Routine

organization members are prevented from promoting interests other than those of the organization, by the existence of a hierarchy between values (common or principal) on the one hand, and actions on the other. There is consistency between what organizational members say and what they do. People at the highest hierarchical level have the right and ability to give commands and issue binding rules. Coordination is achieved by the existence of such local rules, which also produce stability. Organizations have clear boundaries deﬁning who is a member and, via the accounting system, what resources the organization owns. Since every organization has a speciﬁc mission and speciﬁc rules, organizations tend to differ from each other. Missions and rules can also be changed within a single organization, but only from above. This image of organizations make us expect much more heterogeneity and change than on markets. Since local managers can set rules for how their own organization shall work, we can expect different organizations to work in different ways and to change their way of functioning when their managers have decided so. In short, organizations are portrayed in a similar way as modern individuals, as actors (Meyer 1994; Brunsson 2006, Ch 1). They are assumed to be coherent entities with clear boundaries vis-a`-vis other actors; they have special tasks, purposes, and preferences, which control their actions, which are thus intentional and rational. They are seen as having interests, as being able to act and make statements. An older view of organizations as a kind of arena for individuals (Lamoreaux 2004) has been replaced by a view where the organizations are themselves a kind of individual and ‘judicial person’. Although these ideal descriptions of markets and organizations differ greatly, they are also consistent with each other. A market requires actors of the kind described in the ideal organization type, i.e. acting uniﬁed entities capable of selling and buying, with clear boundaries deﬁning ownership and with clear preferences. And such actors are likely to need markets – arenas where they can acquire the resources they need to achieve their missions. So the descriptions complement and reinforce each other. If we believe that markets actually work as described, it is hard to perceive organizations as actors on such markets in any way that differs greatly from the ideal type described. And if organizations actually worked as described, they would facilitate the functioning of the markets as described. Doubts about one description would easily raise doubts about the other.

Beliefs Creating Reform: The Case of Markets and Organizations

27

Like Weber (1924), many scholars have used ideal types as analytical tools only and have not considered them to be empirical descriptions – they have been assumed to capture limited aspects of reality only. But this has not prevented the ideal types from having a strong impact on the way markets and organizations are described in the modern public discourse; they are part of institutionalized belief systems. They have had a deﬁnite impact on academic general comparisons between markets and organizations (Coase 1937; Williamson 1975) and in the long-running debate about planning versus market economies (Dahl and Lindblom 1953), or in the discussion about boundary-setting in companies: whether or not they should incorporate some suppliers or customers into their own organizations (Williamson 1975; Scott 1998). Ideal types are simple, easy to understand, and they spread more easily than complete and detailed descriptions of practice (Abbott 1988, Ch 2; Czarniawska and Sevo´n 1996). And they are ideas that seem to be of high generality, and very general ideas are seemingly more worth learning and using than those describing smaller segments of reality. Furthermore, people may sometimes confuse prescriptions with descriptions. General norms may inﬂuence people’s beliefs about what practice is actually in operation. It is not surprising then that they are also popular among proponents of both the market and the organization institution in ideological debates (Nilsson and A˚slo¨v 1990). The ideal types also provide common norms; it has been argued that both markets and organizations which actually work according to these ideas are more efﬁcient than others (Etzioni 1964; Samuelsson 1964). The idea that organizations should work in accordance with their ideal type has been described as one of the most powerful norms in society (Scott 1998). And when organizations are represented by their leaders, they are typically described roughly according to the ideal organizational type. The types regulate intentions even more strongly: it is difﬁcult to present any other intentions for markets and organizations than that they should work roughly as described in the ideal types. The types form the basis for many formal rules and for legislation (Brunsson, K. 2002). Indeed, the ideal types are part of the way we deﬁne markets and organizations: if many of their characteristics were absent in a speciﬁc case, we would doubt whether it was a case of a market or an organization at all. For instance, scholars sometimes deﬁne systems that are legally markets as ‘non-markets’, as subsumed under a different regime such as

28

Reform as Routine

a network system, if the actual interaction differs very much from the ideal market type (Grabher 1993; Ha˚kansson and Johansson 1993). So in all these respects these ideal type conceptions of markets and organizations are important parts of an institutionalized belief system in Western societies. This is not to say that they form all beliefs about markets and organizations; we are generally able to subsume to several beliefs, even contradictory ones (Veyne 1983; Brunsson, K. 1995). But they are important enough to be used as an example here, and to be referred to as institutionalized beliefs.

Studies of practice Although important as beliefs, the ideal type conceptions seem to have regulated practice on markets and in organizations to a limited extent – at least if we are to believe much empirical market and organization research. A long tradition of empirical studies of market and organizational practice point to radical and routine differences from what is described in the ideal types (Ha˚kansson 1989; Swedberg 1994). Many markets have been described as strongly organized with a substantive degree of cooperation, hierarchy, and power concentration (Richardson 1972; Grabher 1993), of social embeddedness (Granovetter 1985) and of shared meanings (Spender 1989; Hellgren and Melin 1992). Particularly, studies of industrial markets have presented a picture far removed from the ideal market type (Ha˚kansson 1989; Axelsson and Easton 1992). Many such markets are best described as networks of multiple sellers and buyers in complex vertical and horizontal patterns. Relations between the actors in the networks are characterized by a high degree of stability and loyalty. Stability is created, since the market actors are heavily dependent on each other. Adaptation to and by the partner is important. Buyer and seller exchange information and knowledge about products and technology, and about other market actors, and together they modify products and production technology or develop new ones. Resources and goods are created within the network. Deep trust and loyalty can easily develop over time between the partners, based on long-term cooperation, insight into each other’s activities, and a history of common problems and solutions.

Beliefs Creating Reform: The Case of Markets and Organizations

29

Market actors are created within the network. One can only become an actor on this kind of market as part of a network. The preferences of the buyers and sellers are created and moulded in the interaction. Actors are interested not only in their own proﬁtability and viability but also in those of their partners. When buyers and sellers are large organizations it is seldom meaningful to analyse the whole organization as a single actor. Rather, individual departments, groups, or even individuals are the decision units for buying and selling; they are parts of their organizations that can differ from other parts, including top management, as regards both resources and preferences. It follows from this description that different markets may function quite differently depending on the detailed characteristics of the market actors and their relations, and that this functioning can change over time (Callon 1998). As indicated in Chapter 1, the same kind of discrepancy between ideal type and practice is described in empirical studies of organizational practice. Such studies have produced a picture that deviates considerably from the ideal organization type. This is true not only for recent studies of the ‘socially responsible’ organization as described in the preceding chapter. In fact a substantial part of the last ﬁfty years of organizational research have been oriented towards demonstrating this discrepancy (Brunsson and Olsen 1998). Organizations are described as being difﬁcult to control from outside or from above. Instead of being obedient instruments for principals or top management, organizations tend to create their own values and interests (Selznick 1949), or they are permeated by ideas and rules in their environment (Meyer and Rowan 1977). Nor are these values and interests necessarily homogeneous; on the contrary, different departments, groups, and individuals in organizations tend to create and act in accordance with different, conﬂicting interests (Pfeffer 1981). Different departments can act fairly independently of each other and of top management. Control is often exerted from below; top management is dependent on the knowledge of activities that lower levels possess (Baier et al. 1986). What the leaders say and what the organizations do, are not the same thing. Rather than – or in addition to – hierarchy, unity, and consistency, it is local autonomy, conﬂict, and inconsistency that are predominant characteristics of many large organizations (Weick 1976; Pfeffer 1981; Brunsson 2002). Furthermore, internal market mechanisms such as transfer pricing are often installed in large

30

Reform as Routine

organizations. In addition, many of the ideas of how to organize and work in organizations are imported from outside individual organizations, thus creating much more homogeneity and stability in organizations than would be expected if local managers formed their organizations out from their own original and speciﬁc ideas (Powell and diMaggio 1991). Boundaries are described as ambiguous. The interaction between different parts of one organization may be considerably less frequent, less intensive, and less stable than the interaction between one part of one organization and one part of another. When large corporations sell and buy different business units and companies at a great pace, the hierarchical relations in these companies begin to ﬂuctuate a great deal, while external relations with suppliers and buyers may be much more stable (Andersson et al. 1997).

Practice and beliefs These more complex and to some extent contradictory accounts of markets and organizations have been less inﬂuential than the ideal types when it comes to shaping institutionalized belief systems: they are not part of the way markets and organizations are deﬁned; they have little normative impact and they do not regulate intentions. It is hard to intend to create the kind of markets and organizations described in these accounts, at least if one intends them to be ‘true’ and well-functioning markets and organizations. The situation illustrates how our general knowledge in modern societies is to a large extent determined by mainstream science (Meyer 1994; Drori et al. 2003), but also of how relatively simple scientiﬁc ideas spread more easily than more complex ones. The more complex scientiﬁc accounts are attempts at describing practice – descriptions that of course cannot cover all practices or aspects of practices but that we can assume are able to capture some. This means that we will sometimes have a situation where practice works roughly in accordance with these accounts, while belief systems continue to resemble the ideal types. In this case, the differences between the two institutional practices will be less clear than those between the beliefs. Market practice may be fairly organized, and

Beliefs Creating Reform: The Case of Markets and Organizations

31

organizational practice may have some characteristics similar to those of the ideal market type (Friedberg 1993). And more importantly, if we are to believe results from empirical research, it is quite common that institutionalized beliefs and practice do not reﬂect each other in markets and organizations. However, beliefs have other effects than controlling practice. In the next section, I will discuss how the belief systems about market and organization presented above will affect the allocation of responsibility.

ALLOCATION OF RESPONSIBILITY Responsibility is connected to power. In Western culture, power produces responsibility: someone who is believed to be the deliberate cause of events, is also held responsible (Aristotle 1, Book III, Chapter 1; Edwards 1969). It is perceived power, i.e. beliefs about power, that bestows responsibility, not power in practice. As argued above, institutionalized beliefs about the organization portray it as an actor. Organization members are conceived of as acting on behalf of the organization and its management, in a concerted way and with the same purpose. Social actors perform actions and can therefore be held responsible for these actions and for their outcomes, thus endowing them with responsibility. So the organization as a whole is endowed with responsibility. Organizations also allocate responsibility internally, among their own members. This is accomplished by beliefs in hierarchy: some people within the organization are thought to control the others. It is assumed that the leadership of organizations controls organizational activities. Leadership thus becomes responsible, while other organization members become relatively irresponsible. The fact that the leadership’s power in practice may be greatly limited is not important. Organizations often demonstrate, and thus further reinforce, their responsibility and its internal concentration. The standard way for someone to claim responsibility is to claim inﬂuence, to argue that he or she is or has been inﬂuential. It is part of managers’ role to make such claims, for instance by making decisions before or after important organizational

32

Reform as Routine

actions (Brunsson 2007, Ch 8). These claims reinforce the general belief in concentrated power in organizations, thus promoting the further concentration of responsibility in organizations in general. Centralized responsibility is typically reﬂected in legislation about organizations, in which leaders are held legally responsible for a great many things over which they may have had no real inﬂuence or even knowledge (Brunsson, K. 2002). Organizations tend to concentrate responsibility more than power. Institutionalized beliefs about the market, on the other hand, emphasize the lack of concentrated power and thus produce little concentrated responsibility. It is assumed that the actors in the market control only themselves, that they make agreements of their own free will and not as a result of someone else’s power. In this version, nobody has control over anybody else’s actions. Every actor is responsible for their own actions, not for those of others. Nor are actors responsible for the effects of their own behaviour on their partners, so long as they have followed the formal rules of the market. Responsibility is widely spread. Further, the market as a whole is not an actor, and cannot carry responsibility. Nor is any single market actor responsible for the functioning of the market as a whole, or for its results. These limitations on responsibility are reﬂected in most legislation about markets, the few exceptions being clearly speciﬁed. In practice, market actors may have considerable power, as the result of their position in networks or as an effect of access to unique or highly scarce resources. But even when power is concentrated in market practice, institutionalized beliefs tend to obstruct the equivalent responsibility from emerging. It is not only responsibility for daily actions and outcomes that is allocated differently in organizations and markets. Responsibility for the construction and existence of individual organizations and markets is also distributed differently. Organizations are conceived of as human constructions – they are represented as founded and maintained through the efforts of one or more individuals. So it is possible to claim that the founders are responsible for the creation of the organization and that the present management is responsible for its present existence. Typically, responsibility for the existence of markets is more dispersed. When markets are thought of as natural effects of the likewise natural

Beliefs Creating Reform: The Case of Markets and Organizations

33

emergence of sellers and buyers rather than a result of a human plan, responsibility evaporates. And when institutionalized beliefs do not contain strong ideas about how markets are created, responsibility becomes unclear at most. Responsibility for the existence of most markets is considerably less clear and less concentrated than the responsibility for organizations. When beliefs and practice are as they have been described above, organizations tend to produce more concentrated responsibility than concentrated power, while markets tend to produce more power than responsibility. This has practical importance. The different tendencies to produce responsibility may be one reason why people engage in market relations or in organizational solutions. For instance, when managers consider incorporating a supplier in their own organization, or maintaining a market relationship, they can be expected to consider the different effects these solutions will have on their legal and moral responsibility towards the supplier.

BASES AND PROBLEMS OF LEGITIMACY Beliefs about markets and organizations do not only affect responsibility allocation, they also affect the legitimacy of these institutions. Ideas about naturalness constitute one factor: when institutions are perceived as parts of a natural order, they are less likely to be questioned (Douglas 1986). When markets are perceived as natural orders while organizations are not, markets become more difﬁcult to question than organizations. But perceived naturalness does not necessarily make institutions attractive. Although nature is difﬁcult to question, it is not necessarily valued more highly than human constructions. Nature can be perceived as something to avoid, particularly when it is understood as disordered. Market and organization ideas are apt to offer different impressions of order: markets appear to be much less ordered than organizations. As indicated above, organizations seem to be so highly ordered that they are perceived as actors in themselves. Markets, on the other hand, are perceived as arenas only, arenas for organizations and other actors. What

34

Reform as Routine

happens in the market arena is the result of the actions of many actors, and can be quite chaotic.

The individual as a problem for legitimacy The legitimacy of institutions is affected not only by ideas of naturalness and order but also by the consistency between the institution in question and other institutions. If two institutions are invoked at the same time and their belief systems are contradictory, at least one of the institutions is likely to be questioned or criticized. Institutionalized beliefs about markets and organizations are both inconsistent – although in different ways – with some institutionalized beliefs about one of the actors in both systems, namely the individual. The relation to the individual has long been a crucial topic in the moral and scientiﬁc discourse about the market and the organization institutions. The market has been seen as an institution violating norms of altruism for individuals. The assumed tendency of markets to encourage the egoistic preference of individuals for enriching themselves has been a particularly frequent subject of discussion. There is an old debate about whether there should be any difference in the norms for individual behaviour in general and norms for individuals on markets. Aristotle 2 (Book I), Luther (see Segal 1990), and Calvin (see Tawny 1926) all considered the norms for individual behaviour to be no different in markets from those in other areas of life. People should not be egoistic in any situation. Greed and striving for one’s own enrichment were no less blameworthy on the market than anywhere else. And to Aristotle at least the result of greed, i.e. wealth, was equally negative. Mandeville (1724) and Smith (1759) argued against these ideas, claiming that the outcome of collective greed was greater wealth for all. These ideas have been much quoted in contemporary discussions, and institutionalized beliefs about markets assume that market actors are selﬁsh, at least in the sense that they are trying to reach their own goals, and that they are not responsible for others’ ability to do so.

Beliefs Creating Reform: The Case of Markets and Organizations

35

Such beliefs exhibit a certain ﬁt with modern ideas about the individual. The modern person tends to be seen and see himself or herself as a highly autonomous, bounded, coherent, and rational actor possessing a speciﬁc or even a unique set of preferences, which guide his or her actions in preferably a rational way (Luckmann 1967; Meyer et al. 1987). This view of the individual has been much reinforced in modern times: there are more autonomous individuals around, and they exhibit a higher degree of individualism (ibid.). Beliefs about the market are fairly easy to combine with these beliefs about individuals – the market seems to provide an arena where the individual can or should exert his or her autonomy. Modern individuals are apt for the role of market actors: they have the autonomy, the clear boundaries and preferences, which a true market actor should have. The idea of the formal organization, on the other hand, implies the direct subsumption of the individual’s interests and preferences under those of the organization. The preferences of the individual are subsumed under a more general interest, the common task. Individual autonomy and the right to act according to one’s own preferences is reduced by the imperative of organizational hierarchy; otherwise expressed, individual actorhood has to give way to organizational actorhood: if the organization shall be an actor, there is less room for the individual to be an actor. It is difﬁcult to defend this subsumption of the individual when the belief in individualism is strong. So an increase in individualism in contemporary societies can be expected to converge with beliefs about markets, while beliefs about organizations and individuals can be expected to diverge more and more. There have been many attempts at justifying the subsumption of individuals in the organization, and many ideas have been presented whereby individuals and organizations can coexist. Members can be compensated for their participation, thus satisfying their own interest at the expense of the organization (March and Simon 1958, Ch 4). A balance can often be struck which is tolerable to both parties. Or members can democratically elect the leaders they must subsequently obey, thus creating a feeling that to some extent they are controlling themselves. Another notion is that individual organization members have the same interest (and have perhaps created the organization together for that reason). One version of this attempt at reconciling organizational

36

Reform as Routine

and individual interests is expressed in the idea of the ‘corporate culture’, whereby the individuals in an organization share values and beliefs which are also beneﬁcial to the organization (Alvesson and Berg 1992; Martin 1992). Another project with a different angle of approach involves attempts to allow more individualism within the organization, perhaps through job differentiation and advanced accounting systems. This means creating well-deﬁned tasks for each individual organization member, accounting for individual results, and, sometimes, introducing related incentive schemes. Although such schemes have been regarded as a threat to individualism when the tasks and accounting systems are very rigid (Miller and O’Leary 1994), they can be interpreted in the opposite way if the tasks are more ﬂexible and if it is results rather than procedures that are being accounted for. In cultures with a relatively weak sense of individualism, organizations can be expected to be easily tolerated and to have little need for legitimating adaptations to individuals. In such cultures, markets can be expected to be less legitimate. In a somewhat stronger idea of individualism the typical individual may think of himself or herself as a purposive actor, but may be less clear about what his or her purpose or preferences are. This makes individuals weak market actors but might make them willing to commit themselves to a societal or organizational purpose – to become agents for general interests (Meyer and Jepperson 2000). This facilitates organization. If, on the other hand, individuals are not only purposeful but have speciﬁc purposes and preferences of their own and strong claims to uniqueness, then it can be expected that organizations will be difﬁcult to legitimate and markets much less so.

Practical effects of legitimacy The legitimacy of markets and organizations has potential practical effects. It inﬂuences the extent to which these institutions can expand into new areas, or whether their area of application is reduced. Legitimacy also affects the smoothness with which young people can be socialized in the institutions. And again, it inﬂuences the extent to which people follow the rules of the market and organizational institutions.

Beliefs Creating Reform: The Case of Markets and Organizations

37

But the perception of a difference between beliefs and practice within a certain institution can also be expected to have practical implications. The legitimacy of both markets and organizations is uncertain, when beliefs about these phenomena tend to clash with common beliefs about the individual. But if practice is a poor reﬂection of beliefs, it may make the clash less clear. If we ﬁnd the institutionalized beliefs about individual selﬁshness and greed on markets intolerable, we might still accept a particular market in which we have personal experience of people not behaving very selﬁshly or greedily. Or even if we ﬁnd the general idea of the individual’s subsumption under the organizational hierarchy repulsive, we might have no problem in accepting a loose structure and control at our own workplace. Further, the institution of the individual may contain some inconsistency between beliefs and practice, just as the institutions of the market and the organization do. Our beliefs about ourselves and our own actual practice may be far apart, the one having little impact on the other. Individual practice can then be adapted to organizational or market practice, while our beliefs remain unaffected. For instance, even if we heartily subscribe to the general idea of individual autonomy, we may in practice obey many commands issued by higher organizational levels, thus making life in organizations easier. Similarly, a person who subscribes to the Aritstotelian and Lutheran ideas of altruism even on markets, would ﬁnd market beliefs frustrating but would probably ﬁnd it easier to act on markets in practice, not only because markets tend to allow some less selﬁsh behaviour but also because the person might forget some of her beliefs about herself when acting in practice. So, just as markets and organizations differ more as institutionalized belief systems than as systems of practice, both are less reconcilable with beliefs about the individual than with individual practice. This helps to reconcile market and organization beliefs and practice with individual practice. So individuals can tolerate organizations and markets as they actually work in practice, and individuals are seldom an insurmountable practical problem for organizations or markets – even though some people hold beliefs or talk about themselves and others as individuals in a way that is contradictory with their beliefs about organizations or markets. When they think about how these institutions

38

Reform as Routine

generally work, they ﬁnd it almost impossible to combine them, but in their daily practice they can combine them fairly easily. Beliefs are inconsistent but practice is less so. In this situation, the practical aspect of the combined institutions can be described as fairly legitimate, while the belief aspect is not. Practice then continues even if the institutional beliefs are characterized by inconsistency and consequently a certain illegitimacy. A weak connection between beliefs and practice can be an important societal lubricant. To those with little or no practical experience, young people for instance, the inconsistencies between belief systems are likely to appear more serious. For example, if such people cherish a strong belief in the individual, they will regard the organization as highly illegitimate. This is consistent with contemporary trends in society where young people show considerably less tolerance of hierarchy than older people (Inglehart 1977; Andersson et al. 1993). Thus, and somewhat paradoxically, a discrepancy between institutional beliefs and practice may increase the legitimacy of institutions. However, the discrepancy may also lead to another practical effect, the attempts at reform. This is the topic of the next section.

REFORMS A perceived discrepancy between institutionalized beliefs and practices is not always tolerated – it can lead to the questioning of beliefs or the questioning of practice. As argued in the introduction, the questioning of institutionalized beliefs is a difﬁcult enterprise with uncertain effects indeed, where citing one or a few deviating practices as proofs will not have much impact. The questioning of practice can be a much more mundane affair, taking place locally within each individual, organization, or market. The practice of which we have experience may appear not only as an exception in light of our beliefs about individuals, organizations, or markets in general; it can also be regarded as a bad example. This situation provides several reasons for reform, or attempts at changing the local situation.

Beliefs Creating Reform: The Case of Markets and Organizations

39

The local situation may seem to be in conﬂict with institutional norms about the way individuals, organizations, or markets should work; it represents something incorrect that should be corrected. Or local practice does not work as intended. This is what is called a problem, and if it persists over time it becomes a failure. It is a strong incentive to reform. Finally, local practice may not reﬂect the way a ‘true’ organization or market works. This makes criticism of local practice an ontological question: the discrepancy between beliefs and practice makes people doubt whether local practice can be subsumed at all under a certain institution. If relations between buyers and sellers look like those which have been described above as industrial market practice, then people may argue not only that there are ‘imperfections’ or unintended consequences, but even that this is not a market at all. Or it can be argued that an organizational practice like the one described above makes it doubtful whether this is an organization. All these forms of discrepancy between beliefs and practice stimulate reform, i.e. explicit attempts at reducing the discrepancy by changing the practice. It is important to note that such reforms are not initiated by problems within practice but by problems in the interface between beliefs and practice. This explains why success does not impede reform: the fact that an organization is successful does not seem to reduce its reform activities much; there is something wrong anyway (Forssell 1992). But for reforms to occur, reformers are also needed – people who present their intentions and try to implement them. Who become reformers is determined in part by the way responsibility is allocated. Those who are responsible for practice can be expected to react to differences between beliefs and practice, either on their own initiative or as a result of pressure from others. Because responsibility is allocated differently in markets and organizations, the reformers also tend to be different. Because market actors do not bear the responsibility for other market actors, for general market results, or for the formal rules prevailing on the market, they tend not to be reformers. Attempts at reforming markets are normally made by outsiders such as the state or by other central organizations (although of course there may be many market actors who secretly or openly try to exert inﬂuence over these reformers). Attempts at reforming organizations are normally made following an initiative on the part of those responsible for single organizations, i.e. organization

40

Reform as Routine

leadership or their principals. So there are far more possible reformers in organizations than in markets and this in turn affects the frequency of reforms in the two institutions. As argued in Chapter 1, local reforms are in fact typical of large modern organizations. The leadership of large organizations often tries to change organizational structures, processes, or ideologies (March and Olsen 1983; Brunsson and Olsen 1997). Since there are a great many organizations, this means that reform activities are very common, and that a whole industry of management consultants has emerged as a result (Huzcynski 1993). Reform of markets are much more rare. So institutionalized belief systems propel a vast and economically important practice – the reform industry. However, this industry’s effects on day-to-day practice in organizations is less certain. The connection between reform and change is a loose one. There is much change without reforms (March 1981). Even if we see little reforms of markets, individual markets may change dramatically. And reforms in organizations may be plentiful, but they do not necessarily lead to much change because reforms may fail. In fact, organizational reforms often do fail; it proves impossible to implement them as intended (Pressman and Wildavsky 1973; Baier et al. 1986; Brunsson and Olsen 1997). There are intrinsic reasons for reform failure. The belief systems that can be used as the content of reforms are generally unrealistic or at least unrealistic in relation to the practice under reform, which may of course be the reason why the discrepancy between beliefs and practice existed in the ﬁrst place. Moreover, large organizations are generally difﬁcult to control and to change from above, and this is likely to include reform situations.

Reforms reinforcing beliefs Whatever the reasons for reforms may be, institutionalized beliefs tend to shape their content. Reformers present intentions for future practice. It is difﬁcult to present intentions other than institutional ones, or to present intentions that are inconsistent with generally held beliefs of what is a proper and true market or organization.

Beliefs Creating Reform: The Case of Markets and Organizations

41

So, as mentioned in Chapter 1, many reforms of organizations are attempts to make the existing organization more like a ‘true’ one, resembling what we think an organization should be. They include attempts to introduce clear goals or a more clear vision or business idea; to introduce better coordination or more top control; and attempts to increase rationality and efﬁciency. Reforms of markets often represent attempts to install more ‘real market’ principles, such as facilitating market entry, improving competition, and preventing cartels. Reforms of individuals often involve attempts to create more individualism, which parallels reforms in organizations: one may try to ﬁnd out what one’s speciﬁc identity and preferences are, to ‘realize oneself’, to get more of a purpose in life; or to let the mind get more control over the body and start a diet or stop smoking; or one may try to become more efﬁcient and rational, making plans or budgets. Even when reforms do not have much effect on subsequent everyday practice, they may have effects on belief systems. As described, reforms are directed to change when it comes to practice while they are conservative as to beliefs: they are attempts at changing practice according to existing beliefs. They can be interpreted as attempts at making beliefs and practice consistent by preserving beliefs while changing practice. While there are some difﬁculties at achieving the practical part, there seems to be some mechanisms that could help the belief part – mechanisms that can make reform attempts stabilize beliefs rather than destabilize them, even when the reforms do not succeed in changing practice. One such possible mechanism is a difference in the visibility of reform attempts and reform failures. Reform attempts dramatize change: they are often proclaimed loudly and extensively; this is a common tactic for getting reform activities started. Reform failures are not usually represented as loudly – few important people have an interest in revealing the failure. Also, the lack of implementation often appears long after the reform was announced and its relation to the reform attempt may be highly ambiguous. Thus, extensive reform activity in organizations can give the impression that big changes are going on, even if very little change is actually taking place. And reforming managers demonstrate responsibility – they have made the decision to reform. On the other hand, a lack of reforms in markets can give an impression of stability and lack of change. And with no reforms no responsibility is demonstrated.

42

Reform as Routine

The number of reforms that are launched may also reinforce the impression of heterogeneity. If organizations present a variety of reforms over a long time, it is easy to get the impression that organizations are different from one another. If we listen only to the way organizations represent themselves, it seems reasonable to assume that organizations which have announced different reforms are now functioning differently from each other, and that organizations which have not yet announced a particular reform are different from those which have. If most changes in the way market actors function and interact do not need to be, or cannot be, represented as reforms, markets are less likely to be seen as heterogeneous. When differences are not represented in reforms, there is less reason to believe that different markets work in different ways. In this way, for markets the impression that actors and interactions are stable and homogeneous is preserved, even if in practice there may be a lot of change and heterogeneity, whereas for organizations the impression is reinforced that there is much change and heterogeneity in the way they work, even when closer studies would reveal much stability and homogeneity. This in turn may affect beliefs about markets and organizations. For instance, it may lead analysts to believe that organizational behaviour is greatly dependent on local factors in the individual organization, while market behaviour is dependent on the general conditions of all markets. Further, the volatility and heterogeneity of organizations are depicted as the result of the efforts of reformers; this seems to reconﬁrm beliefs that organizations can in fact be effectively controlled by certain people and that these people are highly responsible. Changes on markets, if they are observed at all, do not seem equally controlled or controllable. Such impressions may reinforce the picture of markets as something natural and organizations as something constructed and controllable. So, attempts at making practice more consistent with beliefs may reinforce current beliefs rather than change practice, thus actively preserving the inconsistency between beliefs and practice. Just as beliefs may inﬂuence practice without making the two consistent, in this case the speciﬁc practice when it comes to reforms reinforces beliefs while preserving inconsistency.

3 Constructing Organizations: The Example of Public Sector Reform*

Social science has a tendency to take its objects of study as given, seeing them as natural entities of some kind. Individuals, organizations, states and other basic social units just seem to be there, waiting for the social scientist to study them (Strang and Meyer 1993). Alternatively, however, phenomena such as these can be regarded as highly problematic: why, we might ask, do people believe in them; what meanings do they assume in various social settings; and under what circumstances do they become perceived as real and important aspects of reality? In other words, we can ask questions about the social construction of reality (Berger and Luckmann 1966). Social theorists’ constructions may not always coincide with those of other people; what the theorist sees as a certain type of entity, may be interpreted differently by the practitioner. In other situations, the abstract and general concepts of theorists have a considerable practical impact. Theoretical concepts are based on the study of the actions, descriptions, and interpretations of practitioners. Once formulated, they are often reintroduced later into the world of practice. There they are compared with current practices and used to determine what is good or bad, what is lacking and what needs to be done. Concepts travelling back and forth in this way, between theory and practice, are a * This chapter was co-authored by Kerstin Sahlin-Andersson.

44

Reform as Routine

common feature of late modernity (Giddens 1990). Theoretical concepts are used for developing practice as well as theory. In this chapter, we discuss the social construction of organizations and the way in which this is inﬂuenced by the theoretical concept of organization. The examples will be taken from recent public-sector reforms in several European countries, including Denmark, France, Germany, Great Britain, and Sweden, as well as in Australia and New Zealand. Constructing organizations involves the setting up or changing of entities in such a way that they come to resemble the general and abstract concept of organization. We will argue that traditional public services in many countries have lacked some of the key aspects of organization. They can be described, at the most, as conspicuously ‘incomplete’ organizations. When existing services have been compared with the organization concept, their incompleteness in organizational terms has become obvious, and they have seemed to call for reforms to render them – in this sense – more complete. In fact, many publicsector reforms can be interpreted as attempts at constructing organizations. This interpretation provides some clues as to why the reforms occurred at all, why they acquired their particular content, and how they were received. In other words, our present focus is on the origins and reception of reforms rather than on their implementation or their possible direct or indirect effects on the services offered by the public sector. The next section will be mainly descriptive. We will show how popular reforms can be interpreted as ways of constructing organizations. In the third section, we will discuss the extent to which the concept of organization, as revealed in the reforms, deviates from previous forms adopted by public-sector services, and how it deviates. This helps to explain the content of reforms, i.e. what had been seen as calling for change. In the fourth section, we discuss whether the constructive nature of the reforms can explain why they were launched on such a massive scale and why they gained acceptance so easily. A more fundamental question is why the concept of organization came up at all as a model in contemporary public sectors. In the ﬁfth section, we will offer some speculations on this topic.

Constructing Organizations: The Example of Public Sector Reforms 45 REFORMS THAT CONSTRUCT ORGANIZATIONS Over the last twenty years or so it has been possible to observe a great number of public-sector reforms in several OECD countries (Hood 1991; Olsen and Peters 1996; Olson et al. 1998). In many cases, the reforms were not aimed at the products of the public sector, at health care or education for example. Rather, they have represented attempts at changing the modes of managing, controlling, and accounting for the actual production of such services. Many of the reforms have met astonishingly little resistance and have been introduced at great speed by governments of various political shades (Hood 1991; Olsen and Peters 1996; Brunsson and Olsen 1997). Public services have themselves taken the initiative in reforming and adopting new management techniques. Students of these trends have described them in various ways: as the result of new or liberal ideas and policies (Fry 1984; Aucoin 1990; Pollitt 1990); as part of a shift in ideals away from hierarchy and the state and towards the market and the business ﬁrm (Czarniawska 1985; Czarniawska-Joerges 1994; Humphrey and Olson 1995; Boston et al. 1996; Forssell and Jansson 1996; Easton 1997); as the expression of a growing tendency for rationalization (McSweeney 1994; Power 1994, 1997) or for putting a numerical value on, and accounting for, more aspects of human life (Preston 1992; Preston et al. 1992; Mouritsen 1997; Power 1997); and as a fashion (Abrahamson 1996; Røvik 1996). In this section, we will argue that the reforms can also, or alternatively, be interpreted as attempts at constructing organizations. The reforms can thus be described as a way of turning public services into organizations, or at least into something closer to this than before. Organization literature has been concerned with organization in the general sense of coordinating actors and activities; it has also dealt with the speciﬁc, social entities known as formal organizations, or simply organizations, such as companies, labour unions, or voluntary associations (Friedberg 1993; Sorge 1996; Brunsson and Olsen 1998). Here we focus on organization in this second sense. The literature on formal organizations tends towards exempliﬁcation rather than deﬁnition (Thompson 1967; Scott 1998; for examples see Barnard 1938; March and Simon

46

Reform as Routine

1958), but seems, nonetheless, to provide a fair degree of consensus on some of the important aspects that constitute organizations. Below, we will describe these aspects under the general labels of identity, hierarchy, and rationality and will examine public-sector reforms in relation to them.

CONSTRUCTING IDENTITY Seeing something as an organization means endowing it with an identity. This, in turn, means emphasizing its autonomy, and deﬁning its boundaries and collective resources. Organizational identity also involves the idea of being special, of possessing special characteristics, at the same time as being part of a highly general category, the organization. Many reforms represent an attempt to install or reinforce such features of identity in the public services.

Autonomy An organization has a certain degree of autonomy. It is hierarchically subordinate to only a small part of its environment, for instance its owners. Its autonomy implies that it can own, or at any rate control, resources, as well as controlling its own boundaries, by commanding the opportunities for entering or exiting (Weber 1968; Ahrne 1994). In many countries, attempts have been made to increase the local autonomy of public services, such as schools (Schick 1996), universities (Colado 1996; Easton 1997; Musselin 1997), hospitals (Freddi and Bjo¨rkman 1989; Harrison and Pollitt 1994; Schick 1996; Ham 1997; Lapsley 1997), local governments (Boston et al. 1996; Fernler 1996; Forssell and Jansson 1996), or state authorities (Martin 1995; Hood 1996). Rules emanating from the centre have become fewer and less speciﬁc (OECD 1995); instead, decisions are taken in the relevant local unit. The task of the central bodies is now more frequently framed in terms of giving advice and follow-up rather than imposing directives on local units (Jacobsson and Sahlin-Andersson 1995). Staff is employed by the local

Constructing Organizations: The Example of Public Sector Reforms 47 school, the local university, the particular authority, and so on rather than by the state or the local government, as was usual in the past. Conditions of employment, and to a large extent the division of labour, among professional groups are determined locally rather than by central or professional regulations (OECD 1995; Boston et al. 1996; Schick 1996). Local entities engage in setting the boundaries for their own activities (A˚kerstrøm Andersen 1995). Changes in budget procedures have aimed at giving local units and their managements greater control over resource allocations and operations. Integrated funding has been introduced: how the budget is to be divided among various items such as rent, personnel, and technical equipment is not speciﬁed from above, but is assumed to be decided by the local management (Easton 1997; Olson and Sahlin-Andersson 1998). Some public services have even been allowed to acquire and dispose of income from sources other than central funding, at their own discretion (Schick 1996). In other cases, units have been transformed into organizations that are also legally independent, becoming publicly owned companies or even being privatized (Boston 1995; Svedberg-Nilsson 1999).

Collective resources An organization is held together by its members and by the collective resources that it controls (Ahrne 1994). However, an organization is something more than its members and its present resources: such things can be exchanged, while the organization itself continues to exist. Barnard (1938) described organizations in this context by comparing them to human life. Organization theories have often designated life and the ambition to survive as a purpose and an intrinsic value for organizations (Selznick 1957). In the course of the 1980s and 1990s, public-sector accounting was reformed and accrual accounting was introduced in many places (Olson et al. 1998). Result centres were deﬁned, and assets and debts attributed to them were presented on a balance sheet. The asset side of the balance sheet conﬁrms that a public service is more than just its members. Result centres are treated as the owners of resources, such as buildings or equipment, and are assumed to bear the full costs for these resources and for their own personnel.

48

Reform as Routine

Constructing boundaries An organization is deﬁned by boundaries between itself and its environment. A distinction is made between what is happening inside and outside the organization, and these things are evaluated differently (Selznick 1957). This also means that internal and external relations are assumed to be different in kind; customers are treated differently from organization members, for instance. Organizations are ‘controlled’ and ‘managed’ internally, but they ‘adapt to’, ‘inﬂuence’, or ‘handle’ their environment (Pfeffer and Salancik 1978). In connection with the recent public-sector reforms, local entities have been asked to deﬁne their environments and their boundaries in policy documents. Under extended accounting systems, boundaries are constructed and maintained between the organization and its environment, thus deﬁning costs, revenues, assets, or results as internal or external (Miller and O’Leary 1987; Miller 1992, 1994). Public services used to have pupils, patients or colleagues, i.e. people belonging at least in certain respects to the individual service. These services now deﬁne their own environments in terms of customers and competitors, i.e. actors clearly external to themselves (Forssell and Jansson 1996). More autonomous local bodies have changed their relations with each other and with the central government. They have become providers and customers: they sell services to other public bodies, or buy from them (Boston et al. 1996; Easton 1997). They face competition from others. They may even ‘sponsor’ activities – the Swedish Foreign Ofﬁce ‘sponsors’ the state-owned Royal Opera. Rather than following commands emanating from higher levels, local entities negotiate with such levels and sometimes enter into contracts with them. Swedish universities have established a private cooperative organization to work for their common interests vis-a`-vis the central state.

Being an organization An organization is also deﬁned by conceptual boundaries: it is likely to be presented as belonging to the category ‘organization’ rather than just to some other category such as school, hospital, etc. This has occurred in the

Constructing Organizations: The Example of Public Sector Reforms 49 case of many schools and hospitals, as well as in various state authorities (Sahlin-Andersson 1996; Schick 1996; Mouritsen 1997; SOU 1997). In some British and New Zealand policy statements, it has even been stated explicitly that public services should be regarded as organizations no different from any others (McSweeney 1994; Boston et al. 1996). The base for recruiting leaders and personnel has become increasingly deﬁned according to this broad concept of organization. New leaders for schools, hospitals, libraries, and so on, are not recruited exclusively from other schools, hospitals, etc. Explicit policies have been designed for recruiting public-service managers from other types of organization (Hood 1996; Rombach 1997a), and the fact that there are so few leaders in the public sector with experience from non-public organizations has been seen as a problem (McSweeney 1994; OECD 1995). The experience of leading and managing organizations in general has come to be regarded as an important career qualiﬁcation, more valuable than experience within the practical ﬁeld of the public service concerned.

Being special Identity also involves the idea of being special – the idea that one entity is different from others, that it has certain characteristics or combinations of characteristics that are different from those of the others. The organization has a special task or purpose. It possesses certain features, such as a special competence, special resources, a special structure, or a special way of working; or it represents special ideas. It has its own history or even its own organizational culture. Public services have been asked to formulate their own particular objectives and to produce policy statements and performance reports (Boston et al. 1996; Hood 1996; Power 1997) in which they emphasize the differences between themselves and other services. Individual schools (Jacobsson and Sahlin-Andersson 1995), maternity wards (Rothstein 1994), and universities (Musselin 1997), for example, have developed and projected their own proﬁles. Some have actively marketed themselves with the help of these proﬁles, with logos of their own and sometimes even with new names emphasizing their special identity in order to attract more ‘customers’ (Berg and Jonsson 1991; Rombach 1997).

50

Reform as Routine CONSTRUCTING HIERARCHY

Organizations coordinate activities. In fact, the ability to coordinate action is often seen as the very point of creating organizations (Thompson 1967; Mintzberg 1979). Members of organizations should act in a way that contributes to certain common coordinated actions. Coordination is achieved by an authoritative centre that directs the actions of the organization members. In other words, there is an element of hierarchy within the organization. With the help of this hierarchy, the special identity of the organization is to be transformed into organizational action. The authoritative centre should have some freedom of choice when it comes both to forming the organizational identity and to shaping organizational action. The centre is responsible for the organization and its actions. Constructing such local hierarchies is the objective of many public-sector reforms.

Coordination and control Reformers have tried to strengthen the coordination within the local unit by creating teams drawn from various personnel categories. Coordination and cooperation within hospital clinics, municipal ofﬁces, or state authorities have been reinforced, while spontaneous cooperation among individual professionals representing different units is no longer encouraged (Czarniawska-Joerges 1994; Sahlin-Andersson 1994). The idea is that the internal teams should be guided by organizational policies rather than by central rules or professional norms. This also means that any achievements are attributed to the unit as a whole. Achievements at many universities, for instance, are increasingly attributed not only to the individual researchers, but to the unit, the university, or the research institute. At these universities, management has consequently come to be regarded as an important factor, and university presidents have been given more responsibility (Berrevin and Musselin 1996; Colado 1996).

Constructing Organizations: The Example of Public Sector Reforms 51

Constructing management The authoritative centre of an organization should consist of ‘managers’ who are assumed to exert control and bear responsibility for the organization, its actions, and results (Fayol 1916; Selznick 1957). It is often assumed that, internally, managers are organized hierarchically in layers of leaders and led, whereby the led are leaders of the next layer down (Weber 1968). A board can give general directions to an executive, but the executive is then responsible for organizational activities and results. Several recent public-service reforms have aimed at ‘letting managers manage’ or at ‘freeing managers to manage’ (OECD 1995; Schick 1996). Chief executives are deﬁned as managers enjoying freedom of choice rather than as civil servants following and implementing central directives. These managers are expected to launch their own initiatives, to design activities, and to transform the entity concerned so that it becomes something special, with its own special characteristics. In order to strengthen local coordination and control, many public services have established management teams, and a number of new middle management positions have been created (Humphrey and Olson 1995). As managers have been allowed greater discretion, their leadership role has been emphasized and consequently also the need for training in leadership and strategic management (OECD 1995). Top management is expected to achieve control with the help of management accounting techniques, whereby people or units are made accountable for certain activities and results.

CONSTRUCTING RATIONALITY Another aspect of organization is rationality. Organizations are assumed to be ‘intentional’, in the sense that they work towards speciﬁc goals or purposes (Thompson 1967; Weber 1968). Intentionality represents a kind of process hierarchy, whereby organizational policy is transformed into action. Organizations are also assumed to be rational: executives are expected systematically to forecast goals, objectives, and preferences as

52

Reform as Routine

well as action alternatives and their consequences, and then to allow comparisons between these factors to determine their choice of action. The individual organization, rather than organizational subunits, or society as a whole, is the relevant subject for organizational rational analysis; the relevant objectives, alternatives, and consequences are those related to the organization as a unit. In many organization theories, it has been assumed that organizational actions are governed by decisions under a norm of rationality – a norm that, in practice, is difﬁcult to satisfy. Thus we ﬁnd a lack of rationality being treated in much organization theory as a deviation that calls for a special explanation (Simon 1945; Cyert and March 1963; March 1981).

Setting objectives Public-service reformers have tried to install various systems of management-by-objectives (Christensen 1991; Rombach 1991). The idea is that objectives speciﬁc to the individual entity should replace rules, so that the objectives then control the action, and that the achievement or otherwise of the objectives can then be registered accordingly. Rationality is easier to achieve if the organization has a single, possibly a few, but not too many objectives. Too many objectives make it more difﬁcult to determine what action is the best. Consequently, the demand for rationality in the organization often leads to attempts at simpliﬁcation, reducing the number of objectives, possibly even to one – a single abstract goal perhaps, or a clear hierarchy of one overarching goal and several sub-goals (Cyert and March 1963). A recurrent complaint in recent reviews of public-sector services is that they embrace several tasks and objectives, which to some extent contradict one another. A number of public-sector reforms have been intended to break services down into smaller units, so that each unit could have one clear objective. For example, inspection and service-supplying units or purchasing and providing units, or inspecting and advising bodies have been separated from one another (OECD 1995; Boston et al. 1996; Fernler 1996; Laughlin 1996; Schick 1996).

Constructing Organizations: The Example of Public Sector Reforms 53

Measuring results and allocating responsibility The demands for rationality in organizations call for the rational justiﬁcation of past actions. Objectives and action are thus assumed to be connected in a systematic way. Organizations are expected to account for their actions with the help of the preset objectives; and they are expected to be efﬁcient and to achieve their objectives at minimal cost (Thompson 1967; Ahrne 1994). Accounts are presented to higher levels in the hierarchy, or to external stakeholders with a legitimate interest in the organization. These statements describe the activities and results of the organization in recognized terms, that permit comparison with other organizations (Hopwood and Miller 1994). The results of public-service operations are being increasingly controlled by means of the accounting system or by speciﬁc evaluations comparing results with objectives (Power 1997). Whereas accounting in the public sector was previously regarded largely as a way of reporting revenues and expenses, it is now seen as a device for controlling results (Roberts 1991; Mouritsen 1997; Olson et al. 1998). A disappointing result is expected to trigger some reaction from the top. In some cases, the required results are assigned numerical values and are speciﬁed in such detail that they actually control things as much as the earlier rules ever did (Lindstro¨m 1997). Unlike rules, though, a focus on results passes responsibility to the local leadership, since it is their job to decide the means whereby the results are to be achieved. Managers who are free to choose the means are also responsible for the results of their choices, or, to put it the other way round, if they are to be held responsible, they must have freedom of choice. In the British public sector, the concept of the ‘accountable manager’ is often cited (McSweeney 1994). The accounting system has helped to construct the idea of the rational organization, in which speciﬁc individuals can be identiﬁed as being in control and bearing the responsibility (Miller 1992; Roberts and Scapens 1985). In France, contracts or contract-like agreements have been drawn up between central government and the universities, in which speciﬁc objectives and expected results are given for each individual university (Berrevin and Musselin 1996; Musselin 1997). In Germany, the focus of local government budgeting has shifted away from the input side and is now

54

Reform as Routine

based on the performance achieved by clearly deﬁned result-centres (Lu¨der 1998). In New Zealand, performance agreements covering the performance of the individual manager, as well as the unit as a whole, are made between chief executives and their ministers (Boston et al. 1996). Managers and the units they manage are then monitored every year in light of these agreements, and the chief executive’s pay is set accordingly. In the British public administration system, performance-related pay was introduced in 1985, following what was known as the ‘ﬁnancial management initiative’, whereby aims, costs, and performance indicators were set for the various departments, which thus became cost centres (Hood 1996).

REFORMS IN PACKAGES OR IN STRINGS Reforms in the public sector do not usually proceed by introducing various aspects of organization one by one, as described above. Rather they tend to come in packages or in strings: a particular reform focuses on several aspects of organization at the same time; or reforms focusing on one or a small number of aspects are followed by other reforms focusing on other aspects. Packages or strings of public-sector reforms like this have sometimes even been given their own collective labels. ‘The New Public Management’ is such a tag (Hood 1995). For example, when Swedish pupils were given more opportunity to choose their own school and each school’s revenue was made contingent on its pupil numbers, the pupils and their parents began to take an interest in the special qualities offered by individual schools and, in turn, each school began to invent and emphasize its own particular proﬁle. There was to be more cooperation across professional borders within each school, but there was less cooperation with other professionals outside it; other schools came to be seen as competitors rather than as parts of the same local authority, while teachers and certain kinds of resources became clearly associated with speciﬁc schools (Sahlin-Andersson 1994). Similarly, French universities used to be governed by several ministries and experts associated with various disciplines (Berrevin and Musselin 1996). Budgets were allocated to speciﬁc curricula. After a budget reform

Constructing Organizations: The Example of Public Sector Reforms 55 in 1989, this was changed, and resources were allocated to each university as a lump sum. The universities were asked to analyse their own situation and deﬁne four-year plans, setting priorities between the various curricula and other activities. They acquired a new interest in providing information about their situation, and spotlighting their speciﬁc qualities, all of which served to strengthen the power of the university presidency (Musselin 1997).

Constructing organizations Constructing organizations means introducing certain aspects of organization – labelled above as identity, hierarchy, and rationality – that are constitutive of organizations. The constitutive character means that were these aspects to be lacking in a certain setting, it is doubtful whether that setting would be called an organization at all. Reforms aimed at installing these aspects can be interpreted as a way of constructing an organization, as organizatory reforms. If the reforms appear to be successful, then a reformed entity is more likely to be perceived by many people as a ‘true’ or fully ﬂedged organization, possessing the standard organizational attributes. We have given examples of recent reforms in the public sector and have shown how they assign various aspects of the concept of organization to individual public services. Public services such as schools, hospitals, or state authorities have been constructed as organizations by means of organizatory reforms. Whereas before the state could perhaps be described as a single organization, albeit consisting of many subunits (Ahrne 1998), its construction is now a kind of polycentric network consisting of many separate organizations (Stewart and Walsch 1992; A˚kerstrøm Andersen 1995; Martin 1995; Hood 1996). Also, whereas relations between public entities used to be characterized by many of the typical attributes of large hierarchies, such as setting rules, giving orders, inspecting, and providing information, their interaction now includes features that are more typical of the relations between autonomous organizations, such as competition, collaboration, negotiation, advising, contracting, selling, and buying.

56

Reform as Routine

These organizatory reforms are similar to other attempts at constructing organizations, for instance the creation of a single new business ﬁrm out of two or more previously independent companies through merger or acquisition, although this time, it is a question of creating one organization out of several rather than the other way round. In such cases, an attempt is made to replace separate identities, hierarchies, and rationalities by new, common versions (Nahavandi and Malekzadeh 1988). New organizations are also created by splitting up larger ones, for instance when the vast hierarchies of big corporations are divided over separate business units, or are sold off as part of an ‘outsourcing’ strategy. The recent interest in such measures in the business world represents a slightly delayed parallel to the public-sector reforms, and, in particular, to the more radical examples such as privatizations. In both the public and private sectors, the practical effects of constructing organizations can vary. In merged corporations it is often very difﬁcult, in reality, to get employees to behave as though there were now only one new company (Sales and Mirvis 1984). Likewise, reforms in the public sector often make little impact on the products and services provided, or the impact they make is quite different from what was intended (Pressman and Wildavsky 1973; Brunsson and Olsen 1997). In the organizatory reforms described here, the practical effects are equally uncertain.

DEGREES OF ORGANIZATION: ACTORS, AGENTS, AND ARENAS Despite their constitutive character, the aspects of organization described above should be regarded as variables rather than constants; they can occur to a varying extent and in a variety of combinations in different organizations.

The actor When an organization clearly exhibits all the aspects of organization, it is likely to be regarded as a fully ﬂedged, ‘complete’, or ‘true’ organization.

Constructing Organizations: The Example of Public Sector Reforms 57 Corporations and independent companies in contemporary market economies are often viewed in this way. They can also be regarded as social actors, an actor being an entity with a clear and strong identity, hierarchy, and rationality. Alternatively, using the terminology adopted in Meyer et al. (1987), an actor is an entity possessing independence and sovereignty, with autonomous or self-interested goals, with rational means and qualities, commanding independent resources, and having clear boundaries. In our own culture, these attributes are also attached to individuals. Like the actor-organization, the actor-individual is also assumed to exhibit a high degree of identity, hierarchy (between soul and body), and rationality. An actor is able to act. Both actor-organizations and actor-individuals can perform actions: they can invest, produce, sell, and buy, and they can make statements, and since autonomous actors choose and control their own actions, they also become responsible for them (Chapter 2). The fact that an organization is perceived as an actor possessing all aspects of organization, is no obstacle to the launching of organizatory reforms. Management may want to make minor or even major changes in what constitutes the organization’s identity, hierarchy, or rationality. They are also likely to perceive any contradiction between norm and practice: although the organization is constructed as an actor, it may not function in exactly the same way (Brunsson and Olsen 1997). As a long tradition in organization studies tells us, practice in large organizations seldom lives up to the ideal image: deviations from a strong identity, hierarchy, or rationality are common in practice, which provides an argument for reform, for trying to make the organization actually work in the way it should. Thus, although business corporations are often seen as actor-organizations, they are frequently the subject of organizatory reforms. Attempts are made to reformulate, reinforce, or implement the current business ideas; or to reinforce or abandon what is regarded as established organizational culture; or to reinforce hierarchy by introducing new leadership principles or more intricate management techniques. Reforms intended to increase rationality may involve trying to formulate clearer objectives, to improve the systems for gathering and utilizing information, or, in the old days, to introduce better planning techniques.

58

Reform as Routine

In fact, such reforms reconstruct an existing organization rather than constructing a new one. In other cases, organizatory reforms are more radical, involving the introduction of a new way of conceiving reality. This is obviously true in the case of trying to create an organization from scratch or by outsourcing, and to some extent in trying to create a single organization out of many, as in mergers. Organizatory reforms are also radical in many public services, i.e. in services that have hitherto lacked many aspects of organization, not only in practice, but even in principle. We will now describe two ideal types, namely the agent and the arena, that can usefully capture some of the characteristics commonly found in many traditional public services.

The agent Organizations can be seen as instruments for other actors, for organizations, or for individuals. Companies can be seen as instruments for their owners, subsidiaries as instruments for their parent companies, departments as instruments for head ofﬁce, or public services as instruments for the politicians. Entities of this instrumental kind, working on behalf of others, can be described as agents: they act as agents for their principals. They form an administration or verwaltung. Agents are not complete actors in themselves; the principal is supposed, at least indirectly, to control the action and to be responsible for it. If we regard something as an agent, we attribute qualities to it other than those we ascribe to an actor-organization. An agent has an unclear or weak identity. Resources are controlled by the higher level and are typically allocated through a budget process, where the formal decision is taken about who gets money, and how much. In addition, the higher level may control employment or the terms of employment. An agent also has unclear boundaries. The results which an agent is to achieve may actually occur outside its own borders. The results of an R&D department, for instance, ultimately show up in the sales ﬁgures for the company as a whole. The police are expected to reduce criminality in society at large. Moreover, in order to achieve results, agents often have to cooperate with others. Schools, for example, have to cooperate with the social services and with parents in order to achieve good results. Even an

Constructing Organizations: The Example of Public Sector Reforms 59 agent’s clients sometimes have to cooperate, and it is not clear whether they should be regarded as internal or external to the agent: e.g. are the pupils part of a school, or not? There may be detailed rules that apply to many agents. In many countries, the police departments, the prisons, or the tax authorities have been assumed to have the same structure and to work along the same lines all over the country. Opportunities for the local ofﬁcers are restricted when it comes to inﬂuencing goals, structure, working procedures, or results, which means that their responsibility is also very limited. Responsibility resides mainly with the principal, with the politicians at the top of the state structure, for instance, or with top managers at the corporate level. The survival of an agent is not a legitimate interest. Agents are instruments for solving certain tasks, and have no value in their own right (Weber 1968). Nor is an agent expected to have interests of its own. If an agent actually strives for survival and growth, it is considered to be dysfunctional (Crozier 1964). Agents are supposed to fulﬁl centrally given tasks or to follow centrally given rules, which leave little room for local intentions or rationality. Agents may have to try to fulﬁl several inconsistent objectives at once. The chances of achieving rationality are then likely to be small. The results achieved by an agent are difﬁcult to identify if the agent is able to make a partial contribution only to the achievement of a particular general goal. It is notoriously difﬁcult, for instance, to single out the contribution of the accounting department to the common result achieved by a corporation. Things are even more problematic if the connection between means and ends is ambiguous. The interventions of social workers may help their clients to a better life, for example, but they could also make these clients more helpless and dependent than they were before. Such an uncertainty in the connection between means and ends makes it difﬁcult to use rationalizing models, or to measure the effectiveness and efﬁciency of an individual agent. When results are difﬁcult to measure, rules rather than objectives are more appropriate as a way of controlling agents (Ouchi 1977). Although agents are weak as regards identity, hierarchy, and rationality, most organization theorists would probably still refer to entities of the agent type as ‘organizations’. However, the agent’s status as an organization is fragile, and is sometimes threatened: many people would prefer

60

Reform as Routine

to treat a principal and its agent as a single organization, i.e. to consider the combination – the more complete actor – as the organization (Ahrne 1994). A department is often regarded as being part of an overall organization rather than an organization in its own right. The same is often said of subsidiaries or state authorities too, such that the parentcompany-plus-subsidiaries, or the state-as-a-whole are seen as the relevant organizations. For an agent to become a complete organization, it should exhibit all the aspects of organization: local identity, hierarchy, and rationality. It has to be assigned clear boundaries and to be given greater autonomy and more control over recruitment, members, and resources. It has to have a local management which is granted fairly wide discretion. Within reasonable limits, management should be able to set its own objectives and to be free to make its own decisions rather than just following centrally imposed rules. On the other hand, the agent may already possess some organizational features, such as systems for auditing and accounting, or professional managers as opposed to leaders representing the professions involved in its operations. However, in the new organization, the accounting systems would be more ﬁrmly focused on controlling and measuring the results of the organization’s activities, and the local management would be held responsible for these results. This type of management is likely to have a personal interest in the local results itself, and will probably prefer to work towards a few consistent objectives rather than several conﬂicting goals. Management may seek to motivate organization members to contribute to results by raising their awareness of, and interest in, their organization, producing mission statements and generally emphasizing the great importance of the organization as a unit. In agents, the organizatory type of reform involves more radical attempts at change than in the more complete organizations. They not only try to infuse new content into existing organizational features, or to bring practice in line with principles, but mainly suggest new principles that imply a major change for the former agent. The content of organizatory reforms contrasts sharply with the classic conceptions of bureaucracy and with what are regarded as the traditional problems in such bureaucratic set-ups. In that world, the tendency for agents to become autonomous and self-serving has been regarded as one of the main practical

Constructing Organizations: The Example of Public Sector Reforms 61 problems of the bureaucracy, calling for reform in the opposite direction (Crozier 1964).

The arena An entity can also be ‘incomplete’ in terms of organization because its members, and what they do, are legitimately guided by external interests, values, norms, and standards rather than by an internally generated organizational policy. Entities may be set up to represent and handle conﬂicting interests (Brunsson 2002): parliaments provide an example of this. Other entities may represent speciﬁc competencies, so that entry into the entity and the activities of its members is controlled by professional norms and standards. Hospitals, for instance, must recruit registered doctors, whose professional activities are governed by their common education and the explicit norms of their profession. In some countries, universities have not had the right to choose their own professors; recruitment has been controlled by external experts such as professors from other universities. Such entities can be described as arenas, whose members perform their tasks relatively free from control by the local leadership. Instead, they represent, or are controlled by, external parties. An arena may enjoy considerable autonomy vis-a`-vis higher bureaucratic levels, but far less in relation to relevant external ideologies, interests, standards, or norms. The boundaries between a parliament and the political parties are weak and vague, as are the boundaries between a professionally dominated arena and the relevant professions. Values and rules created outside the arena are crucial. Professional norms and standards make units associated with a particular profession similar to each other: it is difﬁcult for a hospital or a university to deﬁne a clear proﬁle or create an obvious niche for itself, except in relatively simple terms, perhaps by advertising the generous resources it receives for certain specialities, or by claiming a higher quality compared with its fellow organizations. The arena may include some of the elements of organization. It may have considerable control over material resources, and there may be a fair degree of internal coordination, but its external governance often leads to blurred hierarchical arrangements. Traditional hospital and university

62

Reform as Routine

managements, for instance, are restricted in their control over activities. Arenas may contain more than one hierarchy. For instance, some professionals may control activities, while bureaucrats have the economic and administrative responsibility for them. Hierarchies within the professions, on the other hand, may be very clear, but they do not coincide with the boundaries of the arena. In arenas built up round professional activities, the individual professionals – doctors or teachers, for example – generally bear a good deal of the responsibility for their own work and are often the main target for external inspection or complaints from clients. It is the professionals who act rather than the arena as such. The external parties associated with arenas, such as political parties, interest groups, and the professions, tend to have their own distinctive rules regarding appropriate behaviour. Members of these external groupings learn how to rationalize the rules they represent, but their rationalizations are based on general, political, partisan or, professional objectives rather than on local, organizational ones (Abbott 1988). For instance, values about being economical or even efﬁcient seldom loom large among the professionals, which can easily mean that what managers regard as good results and what professionals strive to achieve are by no means the same thing (Abernethy and Stoelwinder 1995). For an arena to become a more complete organization, it therefore needs clear organizational boundaries, local hierarchical control over recruitment and over the activities of its members, a clear idea of its own special mission and characteristics relative to other organizations (in particular those with similar tasks), goals for the organization as such rather than general external objectives, and a local management responsible for achieving organizational results. Entities that are perceived as arenas are often referred to by relatively narrow categories – such as parliament, university, or hospital, for instance. Most organization theorists would subsume such entities under the general category of organizations, but they are marked out by certain peculiarities. Just as in the case of agents, whenever the organization concept becomes the norm, arenas’ weaknesses as regards identity, hierarchy and rationality, i.e. their incompleteness in terms of organization, can provide reason for reform. Here too, reforms that transform an arena in the direction of the organization concept imply an attempt at radical

Constructing Organizations: The Example of Public Sector Reforms 63 change. Recent public-sector reforms have included attempts at transforming services with some of the characteristics of the arena into something more resembling the complete organization. The targets for such reforms have usually been services with strong professional afﬁliations.

ORG A NIZATORY REFORM AS COMPLETION We have argued that many public-sector reforms have introduced more aspects of organization into public services, thus turning them into more complete organizations. In other words, aspects of organization constitute the content of the reforms described above. In some cases, the content of these reforms may have been designed without giving a thought to the concept of organization as such. Reformers and reformees may have been attracted by a certain feature of the organization without relating it in any way to the general organization idea. In other cases, reformers have explicitly referred to the concept of organization in arguing for their reforms, as was illustrated in the second section. In yet other cases, the concept of organization may have been the underlying mental image or frame of reference for reformers and reformees, without there being any explicit reference to it. Well-established institutions, such as the organization, may represent ‘knowledge without concepts’ (Bourdieu 1984: 470): institutions affect people’s way of thinking, talking, and acting without being directly referred to. The content of the reforms may have been inspired by a line of thought built on the concept of organization, and the content may have inspired this line of thought. The concept of organization is thus invoked, but not mentioned. The explicit or implicit role of the concept of organization would explain why so many and such extensive organizatory reforms have been initiated, and why they have been so widely accepted. There are several indications that this may be the case. First, the concept of organization may create a problem and thus the need for reform. There is nothing intrinsically problematic about being an agent or an arena. The agent and the arena are both modern, rationalized institutions, well embedded in established systems such as

64

Reform as Routine

the state or the various professions. It is only when such entities are measured against the organization concept that their weakness as regards identity, hierarchy, and rationality emerges as a problem. In a comparison with the general and abstract concept of organization, entities like agents and arenas seem incomplete – they lack certain aspects that the complete organization possesses. This incompleteness stimulates a desire for reforms that construct an organization. Just what aspects should be added to an entity will depend on the initial state, as described in the preceding section. An entity may lack the same organizational characteristics as the agent or the arena do, or perhaps a combination of both. Second, many of the reforms we have described above have been launched smoothly during a relatively short period of time, and have met little resistance. With the possible exception of the most radical reforms, such as attempts at privatization – which in many countries have been a sensitive political issue – the need for reforms has often appeared self-evident, and the reforms themselves have not required much in the way of supporting arguments (Brunsson and Olsen 1997). This could be due to the invocation and acceptance of the organization concept, the introduction of which represents a switch to a different institution: when people ﬁnd themselves in a different type of institution than they have been in before, they tend to change their norms, beliefs, and behaviour much more quickly and more dramatically than they usually manage to do within a given institutional setting (March and Olsen 1989). Change also proceeds more easily if the new institution – as is the case with the organization – is a well-known one. Once people recognize this institution, they are likely to have similar ideas about what should be done, and this facilitates collective change. Third, the organization concept may explain why reformers have often suggested packages of change, or why reforms of individual aspects have come in strings, in both cases projecting an image of a complete or almost complete organization. The various aspects of organization are likely to appear together or to follow quickly upon one another because they all belong to the organization concept, and this concept and its particular logic determine what should be done: a strong local hierarchy, for example, presupposes a certain degree of autonomy; the measuring and accounting for collective results presuppose clear boundaries and a certain

Constructing Organizations: The Example of Public Sector Reforms 65 degree of coordination; accounting demonstrates results and allocates responsibility; rationality presupposes objectives. It is not necessary to wait to acquire practical experience of one organizational aspect before introducing another one; different aspects succeed each other by the dictates of logic (and therefore quickly) rather than awaiting practice. Even the suggestion of introducing one aspect may be enough to trigger the suggestion of others. In this way, the organization concept can also expand and reinforce attempts at reform by creating more reformers: if reformees accept the idea that their entity is, or should become, an organization, they may become reformers themselves. Reforms can also be demanded and initiated by others: the discrepancy between a public service and the organization concept may be perceived by rank-and-ﬁle members or by external parties who will then generate a demand for reform. When the funding of Swedish schools was made dependent on pupil numbers, local managers asked for more autonomy; when parents were able to choose schools for their children, they asked for information about the speciﬁc proﬁles of the different schools. Sometimes ideas about particular features of organization may trigger the idea of organization rather than the other way round. Policy makers or local managers may have been attracted by certain popular ideas about leadership or management accounting techniques, and have later found out that these ideas or techniques actually presuppose the existence of a complete organization. In arguing for the favoured techniques, they have then found it necessary to claim organizational status, and the concept of organization has then inspired further organizatory reforms. For example, people may have wanted to increase quality and may have thus been attracted by quality standards. The advocates of such standards often emphasize that they are useful for ‘organizations’ (Henning 2000), and the standards are indeed based on the organization concept (Furusten 2000). In implementing such standards, the entity concerned easily comes to be seen as an organization that also needs other completing reforms. Hence, even if no explicit reference was made to the concept of organization, when people were ﬁrst attracted by a speciﬁc organizational aspect, nor was it present even as an implicit frame of reference, it may have been brought to mind by the aspect that had ﬁrst aroused people’s interest, thus triggering further organizatory reforms.

66

Reform as Routine

Finally, the concept of organization as a frame of reference may also explain a certain element of imitation in public-sector reforms. Once something has been constructed as an organization, it becomes vulnerable to further reforms of a kind common in organizations. The idea of being an organization may lead to reforms inspired not only by the abstract concept of organization, but also by the example of other more complete organizations. Once a public service has come to be regarded as an organization, it seems natural to compare it with other organizations, and in looking for improvement, it is tempting to compare it with organizations that are believed to be more ‘true’ or complete. When modern corporations are viewed in this way, we can expect an increasing ‘company-ization’ of the public sector (Chapter 4; Forssell and Jansson 1996): the introduction into public services of whatever speciﬁc ways of displaying organizational aspects as happen to be popular among companies at the time. Some of the reforms discussed here have sometimes been described as organizational fashions, and reformers as the followers of fashion (Abrahamson 1996; Røvik 1996). This interpretation on its own does not explain the content of the reforms – there are, after all, many fashions around. We suggest, instead, that following a fashion is more likely to be the effect, rather than the cause of the organizatory type of reform. Organizations are the entities most likely to be vulnerable to organizational fashions. It is not until an entity has been envisaged as an organization, that organizational fashions become positively and routinely attractive, or even difﬁcult to resist. Reforms that construct the organization make organizational fashions spread more widely. New fashionable models for organizational leadership will be more readily adopted, for instance, if there is already an organizational leadership possessing a certain autonomy and authority. Once organizational status has been envisaged, it is easier to argue for the use of experts, consultants, or training of a kind that target organizations and that tend to reinforce the organization concept. Hence reforms that construct organizations may trigger more reforms of the same kind, and all these reforms tend to be repeated. Introducing the idea of organization does not guarantee that practice will be fully consistent with this concept, and the discrepancies between principle and

Constructing Organizations: The Example of Public Sector Reforms 67 practice tend to stimulate recurring reforms, just as they do in organizations that have long appeared complete.

IN CEN TIVES FOR CONSTRUCTING ORGANIZATIONS We have argued that entities like agents or arenas are likely to be perceived as requiring complementary reforms, if they are supposed to be – or to becoming – complete organizations. We have also noted that the very notion of the organization may trigger or reinforce organizatory reforms. This brings us to the difﬁcult task of explaining why people invoke the idea of organization, why they can become convinced that what was formerly accepted as an agent or an arena is, or should now be, a complete organization. On a very general level, it can be argued that the recognition of certain entities as complete organizations is consistent with a current tendency to perceive individual people or organizations as actors, highly autonomous and self-interested, who interact with other actors on markets or in various negotiations (Meyer et al. 1987). If we wanted to explain more fully and in process terms why the concept of organization is invoked, we would need detailed empirical studies of the way the organization concept and the organizatory reforms have spread into various settings. In this section, we will merely indicate a few directions in which such explanations may be sought. First, explanations can be based on the assumption that intentions, strategies, and plans do matter. Perhaps policy makers and reformers have consciously chosen a strategy aimed at constructing organizations. During the 1980s, public sectors were growing rapidly in many countries, and in doing so they came under strong political attack and were criticized for a number of alleged shortcomings. There was also widespread agreement that their expansion should be restricted, or their size even reduced. Attempts along these lines then produced further complaints. The very size of the sector makes it difﬁcult to control from the top, and bearing responsibility for all this while not having much control over it cannot be a very attractive task.

68

Reform as Routine

In a traditional state bureaucracy, the main responsibility resides with the policy makers, since they are perceived as the only real actors. Whether or not they agreed with the criticisms, policy makers may have found the responsibility difﬁcult to bear and the imbalance between power and responsibility awkward to deal with. It may have seemed wise to share this responsibility with others, for instance with a large number of organizations and their managers. Second, the policy makers may have planned to generate something new, but not organizations as such. They may have planned to introduce markets, and acquired organizations as a side effect. The market was a popular model at the time, and markets or market-like conditions were introduced into various parts of the public sector. This may have stimulated the rise of organizations. A market requires actors (individuals or organizations) who can supply or demand goods and services. For the market to function well, it is important that the boundaries between the actors are clearly perceptible; it is necessary to know who owns what, and how property rights are transferred between actors. Market actors must have a certain degree of autonomy – they should be able to choose what to buy and sell, and who to deal with. It is also assumed that market actors are intentional and rational – that they act in accordance with their own intentions and interests (Brunsson and Ha¨gg 1992). Thus, complete organizations constructed as actors ﬁt better with markets than either agents or arenas. If markets are introduced, we can expect more organizations to emerge. The example of Swedish schools, mentioned above, provides a case in point. Markets may constitute an alternative to a strongly coherent state bureaucracy (Lindblom 1977), but they accord well with state organizations. Third, events that were not part of any grand plan, or that were not planned at all, may have helped to reinforce the organization concept. Fairly small changes in views or practices may have created or supported the need felt for more complete organizations. The customer concept, for instance, could have been such a trigger. It was a concept that was enjoying some popularity at the time of our observations (Forssell and Jansson 1996; Sahlin-Andersson 1996), and it obviously agrees better with a powerful modern conception of the individual than the notion of the client or even the citizen. Customers, like the modern individual, enjoy

Constructing Organizations: The Example of Public Sector Reforms 69 autonomy, have preferences, and possess the right and ability to choose what is right for themselves (Chapter 2; Meyer and Jepperson 2000). A customer normally operates on a market. If public entities start seeing an important part of their environment as customers, or if people in the environment start seeing themselves as customers, then people may also start seeing markets and organizations where there were none before. Another factor which may have helped to give prominence to the organization concept could be the popularity of evaluations and auditing. Evaluations have become common in the public sector, typically targeting individual services (Power 1994, 1997; Rombach and Sahlin-Andersson 1995). An evaluation deﬁnes the unit to be evaluated in organizational terms – its boundaries, results, and responsibilities. It also compares results with objectives. Hence the absence of clear objectives is seen as a major problem. When results are difﬁcult to measure, evaluations often come to focus on the administrative systems that are expected to reﬂect the organization concept. In recent Swedish evaluations, for instance, good social care has been equated with the existence of written policy documents and the presence of well-qualiﬁed managers (Socialstyrelsen 1992; Olsson 1995). Finally, directives appear to have become less popular and less acceptable in the public sector, as in many other social settings, and advice now tends to be more popular. Several national boards in Sweden have become advisory rather than directive: their authority is based on expertise rather than on any right to issue orders or binding rules (Jacobsson and Sahlin-Andersson 1995). In the European Union, much coordination and control is supposed to be effected by way of voluntary standards rather than binding directives (Vad 1998). The substitution of advice and standards for directives can be seen as something that creates more autonomy at the local level, but the very act of providing advice and standards also assumes that the recipient already enjoys some autonomy. Directives can be issued to subordinates and agents, whereas there can only be any point in offering advice and standards to actors possessing a certain autonomy, who can evaluate and act upon them if they meet with their approval (Brunsson and Jacobsson 2000, Ch 9). Clearly, all these factors can be both causes and effects: they may reinforce the idea of organization, and they may be reinforced by it. The various factors may also directly reinforce each other. The idea of

70

Reform as Routine

organization in public-sector reforms did not appear all by itself: in fact, it appeared together with a package of other ideas and practices. In the fourth section, we argued that organizatory reforms themselves come in the shape of a package, not as a number of isolated ideas, or that one idea may bring a whole package in its wake. The same may well apply to the explanations mentioned here, i.e. the factors cited can also appear as packages or strings, whereby one idea or activity is accompanied by – or quickly leads to – the next.

CONCLUSIONS Students of social life and the people they study often share common conceptual frameworks, which means that, to a great extent, they interpret or construct the social world with the help of the same or similar concepts. They learn from each other. Students of organizations have often usurped the privilege of deciding when the organization concept should be applied without consulting those who are acting within, or in association with, these putative organizations. There may, in fact, be a discrepancy between how the people observed deﬁne their own situation and the researcher’s deﬁnition of the same situation. People’s deﬁnitions may vary considerably; they may change either quickly or slowly; and they may not be very predictable, given the present state of our knowledge. Nonetheless, the concepts used by people to construct and make sense of their own reality do have important consequences. For instance, people will behave differently if they see themselves as acting in an organization rather than in some other context, and their beliefs and norms will differ accordingly. Every major concept, such as the organization, the market, the family, and so on can be expected to arrive bearing a whole set of beliefs, norms, and practices. We have sought to illustrate this thesis by demonstrating the way in which aspects of the concept of organization have come to be applied to new areas, and some of the consequences that this has had. We have argued that even seemingly minor changes in the way something is understood – viewing it as an agent, an arena, or an organization – have important implications: certain aspects, issues, or problems emerge as

Constructing Organizations: The Example of Public Sector Reforms 71 more important than before, for example, and this, in turn, can affect the kind of changes that are seen as desirable and acceptable. Two issues call for further exploration by researchers in the organizational ﬁeld: we need to ask what it is that makes people liable to construct the organization rather than any other social form, and we need to know more about the consequences of being an organization. Indeed, these issues are fundamental to the ﬁeld.

4 Politicization and ‘Company-ization’ – On Institutional Afﬁliation and Confusion in the Organizational World In the preceding chapter it was argued that the institution of Organization has spread into more areas of society. Not only private ﬁrms but also schools, hospitals, and football clubs are regarded as fairly autonomous organizations in need of many elaborate and general functions. Their role as formal organizations is often emphasized at the expense of their special role in carrying out a speciﬁc kind of activity such as teaching, curing, or sporting. More and more attention is paid to the concept of the formal organization in general, rather than to more speciﬁc subcategories such as factories, banks, retail stores, hotels, armies or universities. The idea that all formal organizations have much in common is emphasized by organizational experts and by scientiﬁc research and education at business schools and universities. Education in management has become extremely popular by being very broad covering phenomena common to all kinds of organizations. In the ﬁrst half of the last century accounting was the main common ﬁeld, but in recent decades the education has expanded to embrace other ﬁelds of knowledge such as marketing, strategy and organization theory, allegedly relevant to all or almost all kinds of organizations. Nonetheless it can be argued that, within the institution of Organization, there is still considerable variation in the form of organizational sub-institutions with conceptions and rules that up to a point are

Politicization and ‘Company-ization’

73

different. Both social science and legal systems have produced categorizations into a limited set of types. It is common to differentiate between three basic types of organizations or organizational sub-institutions (Polanyi 1968; Sjo¨strand 1985; Forssell 1990, 1992; Ahrne 1994). These sub-institutions can be called the ‘political organization’, the ‘company’, and the ‘association’. The political organization is the institution on which most state and municipal organizations are based; the company is the most important prototype for organizing the production of goods; and the association is the main prototype for trade union organizations, other special-interest groups, and so on. These institutions consist not only of ideas and rules regulating the inner lives of organizations but also organizational environments and bases for legitimacy. They have a considerable impact on how individual organizations are perceived and how they work – institutional afﬁliation matters. However, the impact is far from perfect. In this chapter, I shall argue that individual organizations are generally constructed in a way that combines elements from more than one institution. This creates what can be called institutional confusion. I shall also indicate some possible explanations of such confusion. I shall concentrate on two of the institutions: the political organization and the company. In the next section, I shall try to describe them as they ﬁnd expression in laws, scientiﬁc theories, and the ideas that prevail inside and outside organizations. My intention is to separate and clearly differentiate between the institutions, producing descriptions of what we could perhaps call institutional ideal types. My presentation will necessarily be subjective; from a multitude of ideas and perceptions I shall try to construct two logically consistent types. After this section, I will discuss how actual organizations relate to these types.

THE POLITICAL ORGANIZATION AND THE COMPANY As institutions, the political organization and the company deﬁne not only the nature of organizations and the actors in them, but also the composition of their environment as well as the actors in these environments and

74

Reform as Routine

their particular interests. I will ﬁrst describe how the environments of organizations are constructed in the two institutions, after which I will look at the construction of interior life, and, ﬁnally, at the way in which management and management methods are constructed.

Organizational environments Both the ‘political organization’ and the ‘company’ are envisaged as having environments on which they are dependent. Organizations need legitimacy, i.e. they need to be tolerated by their environments, and they need to acquire various kinds of resources from them. But the environments of the political organization and the company are not constructed in the same way – the actors in them are different. The most crucial environment of the political organization consists of citizens. They provide resources and legitimacy. Characteristic of citizens is the variety of their interests and demands. All or almost all of their demands are legitimate, that is to say they must be considered at least to some extent by the organization if it wants to maintain its legitimacy. The citizen has the right to complain. The organization obtains the support of the citizens through the mechanisms of democracy. Democracy is concerned with procedures and management: through the medium of free elections the citizens are represented in the management of the organization – either directly in the management of an individual organization or in the management of some superordinate body. In the company’s environment, on the other hand, the most crucial actor providing the basic resources is the customer. A customer is an actor with very much more speciﬁc demands than the citizen. As the customers of a company, people expect to be able to satisfy one or more of their interests. The various potential customers of a company may of course differ in their interest and demands, but unlike the political organization, the company does not have to satisfy all these; it can choose to serve certain customers only, to specialize on satisfying certain groups, or meeting certain interests. If the citizens are actors with broad interests, the customers are actors of limited interests – namely, those which the company wants and is able to satisfy.

Politicization and ‘Company-ization’

75

So the company disregards some of the interests of its customers and ignores certain other customers altogether. Still it claims legitimacy. The standard argument for this is competition. The company is legitimate if it is operating under competitive conditions. Competition between companies means that people can choose the companies whose customers they want to be. We could say that the company’s freedom to choose its customers corresponds to the customer’s freedom to choose the company. The customer does not complain, he or she simply takes his or her custom elsewhere. Unlike the political organization, the company’s legitimacy is threatened by monopoly. Procedures and management play no direct part in generating resources for the company; it is assumed that the customer is interested not in such things but only in the company’s products. It is by selling products that the company acquires resources. The organization that surrounds itself with citizens looks less autonomous than one that surrounds itself with customers. After all, customers can be replaced if they do not ﬁt, while the citizens are largely given. The political organization is part of a hierarchy, in which the citizens are the clients or principals. The organization is a ‘public utility’; it exists for everybody except itself. The individual organization does the task assigned to it, but it has no intrinsic value if the task disappears. The survival of the political organization is not an end in itself; rather, problems arise if there is difﬁculty in winding up the organization once its task has disappeared. The company has no mission in this sense; the important thing is that its products are in demand, but if the demand falls the company is expected to adapt the products. It should show no loyalty to old ideas; ﬂexibility is all. Thus, for the company, survival tends to become a superordinate value and goal. Survival and growth are regularly presented in the literature of business administration and accounting as valuable and taken-for-granted goals. It is assumed that the company will strive to favour its own and only its own interests in the way of proﬁt, wealth, and survival. These basic differences in organizational environments produce differences in how organizations handle their environments and how they organize. Let us ﬁrst discuss how organizations try to obtain resources from their environments.

76

Reform as Routine

Financing The company claims legitimacy by arguing that it is capable of producing useful products (goods or services) that are in demand and that are produced under competition or at least under the threat of competition. The political organization claims that its services are useful for everyone in its constituency and that it is under democratic control. These different claims are fundamental in getting resources from the outside. The more detailed strategies for ﬁnancing tend also to be different. The company acquires money mainly as a result of selling its products. The political organization is able to generate money which has no connection with products; a typical form of ﬁnancing is taxation. When the money ﬂows into a central unit in the organization, as taxes do, the problem of redistribution arises: a decision must be made about how much each unit is to receive. The political organization therefore has to have a budget, and budgeting becomes one of its crucial and characteristic activities. In the sale of products, on the contrary, the money can be allowed to accrue direct to those who produce and sell these products. The units involved can be responsible for their own results. Such redistribution as is still required can be effected by the pricing of internal goods and services or by some similar mechanism. The budget is not a necessary ingredient in the company. When ﬁnancing is linked to sales, it is important to emphasize the high quality of the products. And when the company wants to borrow money on the capital market, it has to emphasize how well its operations are going now and are going to go in the future. For this reason, accounting rules have been devised to protect the creditors and to prevent the company from overestimating its economic status. The political organization which ﬁnances its activities from taxes, on the other hand, will tell their ﬁnanciers, i.e. the citizens, how badly things are going now or are going to go in the future: it is poor results that demonstrate the need for more money. Since needs are not being satisﬁed, quality is not high enough, and the money is ﬁnished, more is obviously needed. Individual departments argue their case for more money in the same way. The budgeting process is thus apt to become a powerful demonstration of poor performance, of money spent, and of the need for yet more funds. It provides an opportunity for demonstrating to the citizens that they

Politicization and ‘Company-ization’

77

should give more money, and is thus an important means of ﬁnancing. Partly because of this, the budgeting process does not work well as an internal instrument of top–down control: it tends to increase income for the organization as a whole or for individual departments rather than to support principals or top leaders striving for frugality and efﬁciency.

Organizational forms – action or conﬂict? The two institutions – the company and the political organization – correspond to different organizational forms. Starting from the different environments, we can capture the corresponding ideal organizing mode. The company’s core task is to generate products in the shape of goods and services, which are demanded by a sufﬁciently large number of customers and which due to competition should be produced at a sufﬁciently low cost. The crucial problem is to generate these products by effective and coordinated action. This has important implications for ideas of how to organize. The organization principle for the company is the action organization (Brunsson 2002). The prescriptive literature of business administration in particular has spawned innumerable ideas about how to design organizations in order to generate effective, coordinated action. The authors often write as though they were addressing the company institution, rather than companies in the real world. In a nutshell, what they stress is unity, specialization, and problem-solving capacity. Unity is an important structuring principle: it is easier to produce coordinated action if the actors are united than if they are not. It is assumed that anyone recruited into the organization will support its goals and general aims. Unity is created and sustained by hierarchic structures, for example, or by attempts to impose standardized ideologies. Specialization provides further encouragement for unity and shared conﬁdence, as well as yielding scale advantages associated with a business idea, a market niche, or a product group. And ﬁnally, acting calls for an ability to solve problems, in particular the customers’ problems. All these ideas have had considerable impact on how individual companies are organized. The political organization is quite different. Its crucial competence lies in reﬂecting the citizens’ various and often contradictory ideas. The

78

Reform as Routine

conﬂicting demands to which political organizations are subject have important implications for the ways in which it is organized. The task of reﬂecting a variety of inconsistent norms gives rise also to organizational inconsistencies. Organizations can reﬂect inconsistent norms by creating conﬂict structures, by addressing problems, and by producing hypocrisy (Brunsson 2002). When the reﬂecting of inconsistencies is an important task, conﬂict rather than unity will be the appropriate structuring principle. One way of reﬂecting inconsistencies is to create and maintain a conﬂict structure. The political organization will recruit people who, on their own admission, do not share the norms, values, or ideas which other members of the organization represent. Organizational processes, too, will be different, depending on whether they are geared to action or to the reﬂection of inconsistencies. For reﬂecting inconsistencies, problems are more useful than solutions. It is difﬁcult to ﬁnd solutions that will satisfy inconsistent norms, whereas a problem without any solution can be discussed from all sorts of angles without anyone actually having to resolve it. Many issues become ‘problems’, just because inconsistent norms are involved. Insoluble problems are particularly useful: if, for logical or practical reasons, a problem lacks a solution, nobody can demand that it be solved. Rather, it provides grounds for endless discussions and a multitude of conﬂicting action proposals which, it can be claimed, correspond to the contradictory norms that have to be considered. Decision processes that mobilize action do not reﬂect inconsistent demands very well and vice versa. Organizational action is best served by systematically irrational and solution-oriented decision processes that reduce potential uncertainty (Brunsson 2000). But rational decision making provides plenty of opportunities for reﬂecting inconsistencies. Almost all the norms for rational decision making are useful in this context: formulating problems, specifying goals, considering different alternatives, and describing their positive and negative effects; in all these activities, conﬂicts and opposing ideologies can be laid bare. To the political organization, rational decision processes are an attractive proposition, and they tend to emerge from the conﬂict structure and the organization’s interest in difﬁcult problems. They are less attractive to the company for which action is crucial.

Politicization and ‘Company-ization’

79

Organizational action is prepared, launched, and carried out with the help of talk and decisions. But talk and decisions can also be exploited for external use, for demonstration to the world outside, and for the reﬂection of norms prevailing in the environment. Organizations can reﬂect contradictory norms by introducing inconsistencies into their talk and their decisions. Conﬂicting decisions can be made in different parts of the organization or at different times. In addition, organizations can reﬂect inconsistent norms by systematically creating inconsistencies between talk, decisions, and action, i.e. by producing hypocrisy. They can talk in accordance with one set of norms, make decisions in accordance with another, and act according to a third. Hypocrisy easily arises in the political organization, where it is often difﬁcult to implement decisions due to continuous conﬂict.

Generalization and change For the company aiming to produce effective coordinated action, specialization – concentration on a particular type of action – is a natural way forward. The political organization, on the other hand, has every reason to generalize. It has little cause to distance itself from new ideas in the environment, and in fact ﬁnds it easy to take them on board. It is easy to reﬂect virtually all the ideas of the public, even when these grow increasingly divergent or touch upon new areas. Generalizing not only promotes growth; it also gives the organization greater legitimacy. The more ideas and areas the organization reﬂects, the greater its importance to its environment and to individual groups and persons in that environment. The more sides of the individual and the more values that it embraces, the greater its importance to that individual. In a way, the political organization can be said to spill out over the whole of its environment, complementing the events and actions to be found there with its ideological reﬂections, symbols, and meanings. It thus not only increases its own legitimacy, but also enhances the legitimacy of actions and events in the environment. Like the company, the political organization may ﬁnd it difﬁcult to change. In the company insight and discernment may wither; change fails to materialize because organization members fail to see that it is

80

Reform as Routine

needed. The uniformity of their ideas and lack of rationality may prevent them noticing when or where they have gone wrong. The risk is very small in the political organization. On the contrary, it repeatedly and systematically promotes recognition of the need for change. It cultivates conﬂict, rationality, and criticism of the status quo. But strong conﬂicts and rational procedures may prevent the actions needed for accomplishing change.

Leadership and control – results or procedures In controlling its organization, company management receives good help from the customers since the survival of the organization depends on its being accepted by this group; there will be no fundamental conﬂict between the demands they make on the organization and those made by company management itself. The customers can be expected to have demands regarding product design and, via the pricing system, regarding efﬁciency. If the customers do not make heavy demands, there will still be no immediate threat to the company; it will cope, at least in the short-run, despite low efﬁciency and poor product adaptation. Managements can in principle work on a proﬁt control system: provided only that the different units in the company make a proﬁt on their operations, the company’s survival will be guaranteed. How the economic surplus has arisen – as a result of high efﬁciency, good products, successful sales work, targeting the right customers, or by some other means – is theoretically of no interest to management. The leadership of the political organization has no customers to provide such assistance. It has the citizens, who are to make their voice heard by stating demands from above. It is difﬁcult for the citizens, as it is for the political leadership, to agree among themselves on what contributes a good result. Moreover the citizens – and consequently the leadership as well – are often chieﬂy interested not only in what the organization achieves but also in how it achieves it. What is politically interesting and controversial is generally the means rather than the ends, and the means are thus the more interesting target for control. Typically, control is concerned with rules: the leadership, which the public has appointed, prescribes the procedures and methods which are to be used. This

Politicization and ‘Company-ization’

81

makes it more difﬁcult for political leaders to distance themselves from operations in the way that company managements can. Political leaders ﬁnd it more difﬁcult to decentralize; they are more apt to be held responsible for details in the operations.

P O L I T I C I Z AT I O N , C O M PA N Y-I ZATI O N, AND ASSOCIATIONIZATION Thus far we have looked at a number of differences between two institutions, the political organization and the company. It is more difﬁcult to discover such clear distinguishing features between individual organizations in practice. Organizations are more blurred and mixed than institutions; they generally build on features from several different institutions. Organizations often display a substantial amount of institutional confusion. It is not altogether simple for private ﬁrms, for example, to base themselves exclusively on the company, or for state and municipal bodies to associate themselves solely with the political organization. Individual companies borrow certain features from the political institution, and state and municipal organizations borrow certain features from the company institution. In other words, individual companies are to some extent politicized, while state and municipal organizations are to some extent ‘company-ized’. Indeed, there are tendencies towards more mixed forms, at the same time that politicization and company-ization are both making noticeable strides.

Politicization Politicization means that an individual organization takes on some of the characteristics of the political organization. It includes attempts to anchor the organization in a hierarchy of superordinate and more comprehensive ideas and values, and to address the general public and not only the customers. For large companies today it is not enough to sell products and generate proﬁt. In order to maintain their legitimacy they also have to demonstrate in various ways that they are creating and supporting values

82

Reform as Routine

widely held to be decent, just, rational, effective, and modern; in brief, that the company is part of the modern project of progress and justice (Thomas et al. 1987). Providing proof of this is a task not only for business managements, but also for various kinds of overriding organizations set up by the companies, such as various industry organizations. Their job resembles that of the political assemblies vis-a`-vis the authorities in the public sector. When a company associates itself with civic values, its structures and processes and ideologies assume a new importance. It matters not only what it produces but also how; not only results have to be accounted for, but also procedures. This type of politicization is to some extent an old story, but it seems to have been increasing successively over the last few decades. As well as coming up against a growing number of laws and agreements about organizational structure, such as rules governing workplace democracy, companies are also confronted by powerful waves of fashion. In order to win the respect of share buyers, banks, customers, suppliers, the state or other groups, it may be important at different times to have a decentralized, a divisionalized, or a matrix organization. In the same way, other standards prescribe the processes that organizations should or even must employ, like quality management or budgeting. Companies have also become important producers of ideology; they often put a great deal of effort into proclaiming their numerous positive qualities, that they are efﬁcient, service-oriented, environmentally friendly, or of great public utility.

Company-ization Company-ization means that an organization which originally possessed a strong political element, assumes certain features deriving from the company institution. For example, a state or local government authority begins to regard the people in its environment as customers, or these people begin to consider themselves as such. Increasing the relative size of the fees for the authority’s services is both an expression of companyization and, perhaps, a way of reinforcing the customer image. Another form of company-ization occurs when an authority demands more autonomy, or when such a demand is accepted and supported by

Politicization and ‘Company-ization’

83

outsiders. The idea that the authorities should be less dependent on the political bodies is relevant here, and proposals for introducing management-by-objectives, for example, are also often an expression for this kind of thinking. Since the politicians indicate objectives only, the idea is that the authorities can concentrate on action, on getting results, and that control can be geared to results. Company-ization is also involved when an authority is regarded as possessing a value of its own, extrinsic to its original tasks; thus, when the old tasks disappear or no longer attract adequate resources, it should turn to more popular and more proﬁtable activities.

Associationization Apart from politicization and company-ization, organizations can also become involved in what we could call ‘associationization’, i.e. they adopt characteristics belonging to the third institution mentioned above, namely the association. This time the important actors are members. A growing number of companies are trying to acquire members. You can join IKEA’s Family Club, or the British Airways’ Executive Club, or the Account Club at your department store. Members are more attractive than customers, because they are not so disloyal: they are expected to stick to ‘their’ association. On the other hand, members are normally supposed to make decisions in their own associations, but such ideas are rare in companies. Associationization in the public sector, on the other hand, is generally presented as a form of consumer inﬂuence: it is not citizens in general who are to exert any inﬂuence or the customers who are to choose their suppliers; it is those who actually use the services who are to have a voice. In associations, the members usually make some contribution of their own, and sure enough we now hear about consumer cooperation in the public sector. Perhaps in this way it will be possible to avoid all those citizens, with their conﬂicting interests, who interfere with operations through their elected representatives – but without any risk that the consumers will switch supplier or form new associations.

84

Reform as Routine PROCESSES OF INSTITUTIONAL CONFUSION

Several kinds of processes driven by different causes may lead to institutional confusion. I will point to three kinds of processes that might be used as explanations alone or in combination. They are labelled picking out of individual bits, realizing institutions, and adaptation.

Picking out of individual bits Institutional confusion can be understood as a process in which organizations pick out attractive individual bits from different institutions. Actors may simply imitate speciﬁc features of other kinds of organizations, adding these features to their own organization without noticing or worrying whether the feature might be perceived as inconsistent with other features of the organization (Røvik 1996). Such imitations can be assumed to be dependent on ideas of analogy (Meyer and Scott 1994). Their likelihood is greater the more the imitator and external observers believe the model organization is of the same kind as the imitating organization. Such perceived similarity of kind is enhanced if people make little distinction among different kinds of organizations but consider all organizations to be of the same kind, or if they make distinctions different from the one made in this chapter. The features introduced are not perceived as breaking any internal logic of their own organization. If activities are decoupled in the organization, the potential inconsistency between the imported features and other features of the organization does not necessarily become problematic. Both the political and the company type contain attractive as well as unattractive features. The unattractive features of one type make organizations look for attractive features in the other model. A highly political organization may be attracted by management techniques which focus on results and action. A company may be attracted by methods for dealing with conﬂicts or insoluble problems. The picking out of individual bits can be somewhat opportunistic. Managers exploit those parts of an institution that suit their organization or themselves, and refrain from other aspects which do not. Public

Politicization and ‘Company-ization’

85

authorities want more autonomy and would rather have customers than awkward citizens, but they do not want the competition which also attaches to the company institution. Company managements want to demonstrate their organizations’ public utility and to avoid competition, but they do not want to allow civic inﬂuence. Or companies want members, but have no intention of letting the members vote on the content of the corporate activities like in true associations. State and municipal organizations want consumer inﬂuence, but not to allow their consumers to form their own organizations for the services they want.

Realizing institutions Also, either of the institutional types may be attractive as a whole. This can make organizations strive to become more like one of them, to realize them in organizational practice. In this process they do not necessarily imitate features of other existing organizations. It is the type that is attractive and that should be realized. When successful, such a realization process makes the changing organization more like the institutional type than the organizations that are traditionally afﬁliated with the institution in question. Public authorities trying to become more company-like may at least partly become more like true companies than most private ﬁrms. When an organization becomes more like the institution it is already afﬁliated to, institutional confusion is reduced. But when it tries to become like another institution, confusion increases substantially. Such a change normally takes time and during the transition period the organization will be in a confused state. In addition, the transition may never succeed completely; sediments of old features may persist for a long time, maybe because people like them to. This may make the confusion quite stable. There are several incentives to change towards the political type. The weaker autonomy of the political organization can make it appear more valuable in the eyes of its environment. Its ambition to service everyone, to be a public utility, does appear at least to an outsider more worthy than a greater emphasis on self-interest such as is shown by the company. Also the struggle for survival can provide a powerful incentive to transition towards the political type. As a survival strategy the reﬂection

86

Reform as Routine

of inconsistencies is better than the creation of effective action. According to studies or organizational lifespans, the longest-lasting organizations are universities, churches, and states. Organizations whose legitimacy is based mainly on action, however, have a remarkably short average length of life (Starbuck and Nystrom 1981). Specialization may help proﬁts, but it also makes an organization vulnerable: niches can disappear and rapid adjustments are frequently called for. Generalization is a less risky way of growing. Insoluble problems normally stay around longer than problem solutions. More tools for making adaptations are available to organizations with a strong political element. Thus, in terms of survival, it is more sensible to base legitimacy on the reﬂection on inconsistencies than to base it on action. A company will be particularly prone to politicization if it has proved unsuccessful in its ﬁeld of action, for instance if it cannot sell its products at proﬁtable prices, and therefore has to seek the protection of state subvention. In extreme cases of politicization, severe ﬁnancial problems can disappear. Financing acquires a new meaning. For example, in 1988 the Swedish Railways, an organization with long experience of mixed institutional afﬁliations, was split into two parts: one more clearly resembling a company, with the task of running the trains; and the other a state authority, with the task of building and maintaining the tracks. Swedish Railways used to receive substantial ‘subsidies’, and much of this money was now transformed into ‘allocations’ to the state railway authority. In this way, an old problem – that the state had stooped to subsidizing an organization which was regarded at least partly as a company – was elegantly solved. Now, that problem ceased to exist; making allocations to an authority is a perfectly normal thing to do. On the other hand, the ﬁnancing strategy of the political organization makes it appear to be functioning less efﬁciently than the company; constantly reporting poor results, needing more funds, and consuming our money makes an organization less attractive compared to organizations which appear to be producing excellent results now and in the future, and which generate their own funds. The company type is attractive for other reasons too. The greater autonomy of the company, its more unproblematic existence, and its greater energy may all appear attractive. Another reason may be that the concept of ‘money’ has different meanings in the company and in the

Politicization and ‘Company-ization’

87

political organization. In the company it is regarded as something that is ‘created’: the company can ‘generate proﬁt’, and the company is proud of it. High turnover is also a positive phenomenon. The political organization, on the other hand, is regarded as ‘consuming’ other people’s money, and this consumption has negative connotations: the greater it is, the worse it is. If the political organization can show a surplus of money, this is seen in turn as an expression of overtaxation rather than as something which the organization has created itself. All this means that the political organization may be regarded as ‘draining’ and the company as ‘sustaining’. The organizational forms also project different images. The political organization certainly promotes the intellect: its members acquire a great deal of experience in analysing and discussing difﬁcult problems. But a political organization is also ‘depressed’, in the sense that it has to tackle problems which it cannot solve, it creates uncertainty about the nature of the situation and what should be done, and it projects the idea that much of what it does is wrong. Nonetheless, those who work in the organization ﬁnd it difﬁcult to change their own or the organization’s behaviour. And criticism is one of the organization’s important activities. People who like action and assertiveness more than intellectual discussion might want to change a highly political organization into something more like the company. The company’s action-oriented organization, on the other hand, encourages work on solutions and emanates a strong belief that what it is doing is right. These qualities make the organization appear to be functioning more effectively, but allow its members less opportunity to display intellect and different opinions. People who prefer the latter qualities may work for politicization. Managing company-like organizations is attractive since the managers have greater autonomy, better opportunities for distancing themselves and rising above the trivialities of practical operations, and simpler methods of control than in more political organizations. Managing more political organizations, on the other hand, gives room for a more human, intellectual, and good-hearted style and impression. Different qualities ﬁnd favour at different times. Sometimes one type of institution plays the hero’s part, sometimes another. In the 1930s and 1970s the political organization was the most popular; in the 1950s

88

Reform as Routine

and 1980s the company became the hero. This may make organizations swing back and forth between the two models, thus being in transition and therefore confused most of the time.

Adaptation Institutional confusion may also arise because organizations adapt to changes in their environments or to other changes in their own behaviour, which challenge their institutional afﬁliation. For example, one reason for politicization may be that the state now regulates an increasing amount of the operations of business companies – or can be expected to do so unless ‘voluntary’ regulation is undertaken. Another cause of politicization could be what has been called the crumbling of the markets (Hernes 1978). State regulations and efforts by the companies themselves, as well as various technological and economic factors, have led us far away from free competition. This applies not least to domestic markets. When it is difﬁcult to claim that products can be sold, that efﬁciency can be guaranteed, and proﬁt created, all as a result of the customers’ free choice between a large number of independent and non-collaborating companies, it is also important to demonstrate that companies are efﬁcient because they have the right structures and processes and ideologies. For example, mergers reducing competition on the domestic market are sometimes motivated on grounds of maintaining the national industry’s international competitive strength which will be to the beneﬁt of a larger unit, namely the country. Companies that are expected to satisfy a broader range of demands will often ﬁnd that the demands are in fact contradictory. For example, companies are to be proﬁtable but they must also provide employment and a good working environment – and all this without polluting the environment. The existence of such inconsistent demands can mean building conﬂict into the corporate structure, perhaps by introducing co-determination; decision processes may also become more formally rational as they have to cope with a multiplicity of alternatives and contradictory information. Inconsistent demands may generate hypocrisy: a company which cannot quite meet the norms in a certain area is forced to publish ambitious goals.

Politicization and ‘Company-ization’

89

Politicization may also result from an adaptation to another important institution – the individual. When people consider themselves more and more as autonomous individuals with unique preferences and qualities they are more difﬁcult to adapt to the collective and coordinative features of the action-oriented company. Company-like organizations may meet individualization by politicization. Just as politicization can be encouraged by the crumbling of the market, so may company-ization be encouraged by shortcomings of democracy. A fossilized party system, great power to the parties and little to the electorate, powerful corporatism and weak popular commitment to the democratic institutions – such factors make it difﬁcult for the state to sustain a high level of legitimacy with the help of the political bodies. There is then less support for the authorities from above, which makes it more important for them to create legitimacy at their own level instead. Energy and efﬁciency may, to some extent, replace democracy as the grounds for legitimacy. Another reason for company-ization in public sectors in Western Europe may be these sectors’ own enormous expansion. It is simply more difﬁcult to claim that the political organization is a ‘better’ model than the company, in view of the wide range of services and goods for which public sectors are now held responsible. But if the aim is still to retain the monopoly and the illusion of political control, then the company institution cannot take over altogether. The result is all too easily a kind of mixed form, the company-ization of the public administration.

WEAK INSTITUTIONS Thus, in practice, organizations can borrow ideas from several institutions. It is possible to combine customers with political control, companies with membership, etc. This reveals a certain weakness in the institutions themselves. The standards adopted by an organization, its patterns of action, and its environment are not regulated exclusively by any one institution. We who do not belong to the organizations seem to accept that they combine features from different institutions. This makes

90

Reform as Routine

it easier for their leaders to decide how to develop the organizations and how these and their environments are to be perceived by others. Private individuals who feel that their interests differ from those of the organizations would beneﬁt from a debate about the various institutional qualities: which of them belong together, and which do not? Such a debate could clarify the conditions which we would like to see fulﬁlled – in terms of democracy, competition, or members’ inﬂuence – if we are to continue to grant to the formal organizations all the power, the wealth, and the legitimacy which they enjoy in our society.

5 Organizational Reforms as Routines In this chapter, I shall argue that reform can be regarded as part of organizational stability rather than of organizational change. Reforms are often presented as dramatic one-off changes and they may sometimes lead to changes. But reform in itself is more often a standard repetitive activity. Reforms are routines rather than breaks in organizational life. There are many studies indicating that reforms are difﬁcult to implement (Pressman and Wildavsky 1973; Hanf and Scharpf 1978). I shall point out that it is fairly easy to start reforms. Indeed, the very ease with which reforms can be launched may be one of the reasons why so few are ever concluded. It is not an uncommon argument among reformers that reform is necessary in order to adapt the organization to important changes in its environment. I shall argue that several fairly mundane, common, and stable aspects of internal organizational life, as well as various stable and common features of organizational environments, also tend to produce reforms. One of these factors is reform itself: reforms tend to produce further reforms. The argument will be based on certain attributes of administrative reforms* and certain features of organizations, none of which are necessarily universal but which are common enough to be useful in many cases as explanatory tools. There are three common basic attributes of administrative reforms which, together with certain organizational characteristics to be discussed below, make it easier to initiate and pursue reforms in organizations. One is simplicity: the ideas proposed in the reform are less complicated and more clear-cut than most organization members believe the current

*In this chapter the term ‘administrative reform’ is used as a synonym of ‘organizational reform’.

92

Reform as Routine

organizational practices to be. Reform ideas are principles rather than detailed descriptions, theories rather than perceptions, i.e. reforms present ordered ideas which cannot include all the complexity of the real world. Secondly, reforms are normative: they represent attempts to bring order into a chaotic reality rather than to report upon it. Thirdly, reforms tend to be one-sided; the reform tends to invoke a single set of consistent values and perceptions of the world, in contrast to organizational practices which often have to deal with inconsistent values and perceptions by falling back on inconsistent talk and actions. A fourth characteristic of reform is its future orientation: a reform is a process of elaboration, persuasion, and implementation rather than an instant action, so it cannot be expected to produce immediate results. Instead it promises future beneﬁts, sometimes when implementation has been completed and sometimes much further off in the future. Since reforms contain ideas about problems and solutions, they are dependent on an adequate supply of both these things. The supply will be better equipped to nourish many reforms if each problem and solution can be used several times over, as forgetfulness allows. The supply of problems, solutions, and forgetfulness in organizations will be discussed below.

THE SUPPLY OF P ROBLEMS Reforms beneﬁt from problems. The perception of problems in the present functioning of an organization can initiate the search for reforms and offer a strong incentive for attempts to implement them, as well as providing arguments for convincing those whose support is needed. The supply of problems tends to be good in modern organizations (Starbuck 1983). Apart from all the problems that a changing or malevolent environment can generate, there are many internal sources of problems. Administrative reforms can be suggested as a remedy for almost any kind of problem such as low proﬁtability, growing competition, or bad leaders. Administrative reform, it can be argued, will make the organization more efﬁcient, proﬁtable, and more market oriented, or it will make the leaders function better. But administrative reform is regarded as a

Organizational Reforms as Routines

93

particularly natural remedy when the problems are perceived as being administrative, when they are directly concerned with the internal functioning of the organization. The supply of administrative problems also seems to be good in modern organizations. One important source of administrative problems is the tension between the way an organization is presented and the way it actually works (Brunsson 2002). Organizations are often presented to the outside world by their leaders in terms of the institution of the individual. Just as it is an important trait of Western culture to describe human beings as individuals, it is popular to describe organizations using the same ideas. The institution of the individual includes a theory of coherence, consistency, action, and control. The organization is described as working towards a single goal or a set of consistent goals. It is described as a system for producing action in accordance with these goals, something which is guaranteed by management’s control over those who produce action at the lower organizational levels. The theory of the individual is often used in a normative manner, as a way of describing intentions. Even if organizations are not described as working in this way, managements often describe themselves as striving for consistency, action, and control. In doing so, they can use the ideas provided by a vast normative literature of organizational management. Indeed, it would normally be difﬁcult for a management to present itself as striving for inconsistency, or seeking to avoid action or control. However, organizations tend not to live up to these descriptions. The literature describing actual organizational processes tells us that organizations are often characterized by a higher degree of inconsistency, by greater difﬁculty in mobilizing action, and by less top–down control than is ever pictured in organizational presentations or desired by management (Weick 1969; Cohen, March, and Olsen 1972; Brunsson 2000). Even if the theory of the individual may capture some aspects of the lives of organizations and human beings, it also fails to describe important parts of their being. But the tension between presentations and intentions on the one hand and experiences of actual behaviour on the other provides an incentive for repentance, for reform. The theory of the individual describes structures, processes, and ideologies, but not results. Even in organizations which experience success in their results, the tension between the individualistic ideas and actual

94

Reform as Routine

behaviour may be strong. Thus, even in successful organizations there can be a widespread feeling that the organization is not working in the way it should, and that reforms are needed. For instance, good results do not exclude poor control. Management may be aware of big control problems even in successful organizations. Success may depend on the fact that the organization adapts easily to changed circumstances in its environment – that it is ﬂexible. But it has been suggested that organizational characteristics which lead to ﬂexibility may also inhibit opportunities for manipulating the organization in a speciﬁc direction: ‘changefulness’ is not the same as ‘changeability’ (Brunsson 2000). A ﬂexible organization may be successful in a changing environment, but it may also be particularly difﬁcult to steer in the direction that its management wants. Problems are a perceptional category. Some observers may perceive a reality as problematic, others not. Thus, anyone wanting to pursue reforms can try to inﬂuence other people’s perceptions of today’s reality. But it may be easier to refer problems to an area in which we are not troubled by knowledge, namely the future. This suits the future orientation of reforms; since reforms take time, there is a better chance of ﬁnding problems for them. Naturally, the future is especially important to reforming organizations which ﬁnd it difﬁcult to justify reforms on the grounds of present performance, i.e. to well-functioning and successful organizations. An organization’s supply of problems is threatened in so far as it solves the problems. It is therefore good to have problems that cannot be solved. Organizations grapple with several problems which in practice are insoluble. These include problems that involve conﬂicting demands, when it is impossible in practical terms to ﬁnd any balance that could readily be regarded as exactly the right one. How, for example, can the organization strike a balance between the need for integration and the need for differentiation, between centralization and decentralization, between internal markets and hierarchies, or between production and market concerns? Any solution along these dimensions is susceptible to criticism for failing to satisfy one or other – or both – of the needs sufﬁciently. For instance, centralized organizations tend to generate complaints about insufﬁcient consideration of local knowledge and local needs for adaptation, while decentralized organizations will discover that they are not paying enough attention to the beneﬁts of coordination and standardization. And organizations that have struck some sort of balance between centralization and

Organizational Reforms as Routines

95

decentralization may well face complaints from both sides. All these complaints can be used as arguments for reform. Another source of problems which can be used to justify reforms consists of the reforms themselves. Reforms may sometimes solve problems, but they may also trigger problems. The successful implementation of a one-sided reform may well solve a particular problem, while at the same time creating, reinforcing, or drawing attention to other problems which it was not aimed at. For instance, successfully implemented reforms geared towards industrial democracy could produce new reforms aimed at increasing efﬁciency. Reforms along the dimensions of the insoluble problems mentioned above provide another set of examples: decentralizing a centralized organization, for instance, is a way of making sure that the beneﬁts of centralization will be discovered and sooner or later attended to. Reforms may leave the organization oscillating between different solutions. Reforms can also generate claims for further reforms aimed at the same problems and propagating the same set of values or perceptions. This is the case when reforms generate more hopes than they fulﬁl, or even than they could ever fulﬁl. Reforms may be more efﬁcacious at raising the level of aspiration than at improving the situation. Reforms turn people’s attention to the problems to be solved, and the process of selling and implementing the reform can help to raise their aspiration level, making them more eager to ﬁnd solutions and to adopt higher standards about the kind of solutions they are prepared to accept. At the same time, it is often difﬁcult to implement the reform. Or, even if it is implemented, it may not fulﬁl its goals as well as was once expected. For instance, a reform intended to promote democracy may be launched just because the organization is extremely undemocratic. But the reform may lead to a higher and more widespread evaluation of or preference for democracy rather than affecting the actual level of democracy in organizational behaviour – thus demonstrating the lack of democracy and the need for further reforms all the more clearly. In addition, reforms tend to be systematically oversold by their proponents (Baier, March, and Sætren 1986) who tend to promise more than the reform can actually fulﬁl. If overselling is an efﬁcient technique for getting reforms accepted and implemented, we would expect that reforms which have been accepted and implemented will also have been

96

Reform as Routine

heavily oversold. And overselling may help to raise aspiration levels even more. So whether or not reforms are perceived as successful, they tend to lead to the perception of problems for which new reforms are needed. Reforms may trigger a constant ﬂow of problems and consequently the continuous need for reform, so that reforming becomes a stable state.

THE SUPPLY OF S OLUTIONS Problems are not enough to trigger administrative reforms. A supply of ideas for solutions is also needed, solutions which deal with organizational structures, processes, and ideologies and which differ from the solutions presently practised. Solutions, like problems, can be fabricated by those who wish to pursue reforms; but the task of the reformers is easier if a supply of more or less ready solutions is available. Solutions can exert an attraction on those who pursue reforms and on those who are affected by them. Solutions, as well as problems, can obviously provide incentives to reform (Cyert and March 1963; March 1981). Students of organizations are an important source of solutions. There are many theories and ideas from Fayol (1916) onwards which argue that organizational form is important to management control and performance, and which also specify the forms that are best in different situations (Burns and Stalker 1961; Woodward 1965; Williamson 1981). From this long discussion it is possible to pick out ideas for reform, and to provide them with elaborate supporting arguments. But in order to carry out reforms, it is not enough to have solutions. The solutions must also seem better than the solutions currently in use. It follows from the previous discussion that it is easy to ﬁnd these better solutions when designing reforms. Today’s practice may have solved some past problems, but it tends to generate others. These are the problems we are aware of now, or which we will soon be aware of, and it is these which reforms can solve. Thus, unlike today’s solutions, reforms promise to solve the problems of today or tomorrow. And if practices are a bit one-sided due to an earlier one-sided reform, then a new one-sided but reverse reform can seem very promising. If current organizational life is perceived

Organizational Reforms as Routines

97

as very complex, inconsistent, and difﬁcult to understand, the solution of a reform can offer more simplicity and consistency and can be much easier to understand. The reform solution may come closer to the presentation model of the organization, to the way the organization is presented by its management and the way people think organizations should be. If we set a simple, clear, and good reform idea against our knowledge of the current situation with all its slack, ad hoc solutions; and its uncertainties, inconsistencies, conﬂicts, compromises, and complicated relationships, then there is a good chance that the new solution will appear better. It is easier for simple principles to attract enthusiasm and support than for complex descriptions of reality to do so (Jo¨nsson and Lundin 1977). If a reform is to be possible, it is important not only that solutions exist and that they tend to be better than current practices, but also that they exist in considerable diversity. If there were only one best administrative solution, there would be no need for further reform once this solution had been installed. A diversity of solutions is provided in several ways. The theories that describe the causal effect of organizational form on control and performance have two characteristics that are important in providing diversity: they do not agree with one another and they are not completely true. Since they do not agree, the individual organization can draw on different theories at different times and can thus motivate reform. Since in reality there is no revealed simple relation between organizational form on the one hand and control and performance on the other, we will ﬁnd successful organizations with different organizational forms. Consequently, if an organization wants to imitate other successful organizations, it will usually ﬁnd a number of solutions to choose from. Something that both reduces and increases the diversity of solutions is fashion. There seem to be fairly strong fashions in the organizational world, as regards the right solutions and the right problems. These fashions are spread by management books, magazines, courses, and consultants. Fashion reduces the number of acceptable solutions at any one time, but it provides diversity and a strong incentive to vary the solutions over time. Strong fashions guarantee that the practices of an organization will at least sometimes appear old-fashioned and in need of reform. Management consultants may spread fashions, but they also provide some diversity within a particular fashion. If their main means of

98

Reform as Routine

competition is to offer slightly different solutions, they will be eager to create variety in the supply of both solutions and problems. To increase their competitiveness, the less successful consultants at least may seek to update their product. This creates new solutions. Administrative experts within organizations can assume a role similar to that of the consultants. Their task is to reform, to be able to indicate new organizational solutions as well as organizational problems. Consultants and professional reformers, as well as the very idea of reform itself, will of course beneﬁt from the general culture of modernity, whereby change is considered natural and is often deﬁned as ‘progress’, i.e. as leading irrevocably to higher forms, while stability appears backward or unnatural (Berman 1982).

THE SUPPLY OF F ORGETFULNESS Administrative reforms are often repetitions of earlier reforms. The ideas they contain may be roughly the same as in the previous reform, perhaps because the previous reform did not close the gap between aspiration level and practice and was therefore deemed unsuccessful. Reforms may also be repetitions because they are part of an oscillation process. The ideas may be the opposite of those contained in the last reform but the same as those in an earlier reform, whose shortcoming the most recent reform was supposed to correct. In a study of organizational reforms in the Swedish State Railways over a 100-year period, for instance, it was shown that certain problems and solutions were repeated in a great many reforms. A recurrent problem was that top management dealt with too many detailed issues, and the reform implied that they should not do so. Other issues turned up in roughly every second reform, so that reforms about centralization, for instance, were followed by others about decentralization and so on (Brunsson, Forssell, and Winberg 1989). Such repetitions may threaten the initiation and pursuance of reforms. Suggestions about reforming an organization in the same way as on a previous occasion may encounter criticism: the earlier reform obviously did not succeed since we have to do it again, and if the earlier reform did not succeed why should this one? There may even be cynics in the

Organizational Reforms as Routines

99

organization who have experienced so many reforms that they have become sceptical about the very idea of reform as a way of solving problems or improving performance. So reforms are facilitated not by learning but by forgetfulness, by mechanisms that cause the organization to forget previous reforms, or at least those with a similar content. Reformers need a high degree of forgetfulness to avoid uncertainty about whether the reform they are promoting is a good one, and forgetfulness also helps people to accept reforms. Reforms focus interest on the future rather than the present, and forgetfulness prevents the past disturbing the future. There are a number of mechanisms promoting organizational forgetfulness, thus also promoting reform. One example is high personnel turnover; another is replacements in top management. A third is the use of consultants. Consultants – especially if they are different from the last ones – can see the organization with fresh eyes and can thus more easily repeat old mistakes. Management consultants are generally expert at introducing reforms, but they are also normally too busy and too expensive to be involved in implementing them fully. When consultants are used in this way, they tend systematically not to learn about the difﬁculties or the lack of implementation, or about the results of reforms. They are therefore in a particularly good position for initiating and pursuing the same reform in new organizations with great enthusiasm and drive. Finally, reform constitutes a forgetfulness mechanism in itself. Reforms tend to be very different ex ante and ex post. At best, a reform proposal consists of a description of an expected future reality. Descriptions have to simplify and, as we have already noted, the descriptions associated with reforms tend to be greatly simpliﬁed; reforms are launched and sold in organizations mainly in the form of attractive principles. Once a reform has been carried out to some extent, it tends not to be quite so simple and beautiful. The pursuance of reforms is usually characterized by a growing degree of complexity, when one tries to apply the simple principles to organizational realities. As complexity increases, the difﬁculties, conﬂicts, and ‘practical problems’ become more evident. The implementation of reforms tends to undermine them by making it obvious that the original ideas that once seemed so attractive are not clear, precise, or detailed enough for use in practice, and that making them practicable also makes them much less beautiful.

100

Reform as Routine

So just as comparison with today’s chaotic reality makes the simple principles of a reform seem attractive, an old reform under implementation will be less attractive than a reform that is new and still untried. A comparison between one reform ex post and another reform ex ante thus works to the advantage of the ex ante reform, even if the two reforms actually contain the same general ideas. Reforms tend not to deliver what they promise. But their promises are so good that people are easily lured into trying again. To sum up, there are several reasons why we can expect administrative reforms to be common in modern organizations. There is generally an ample supply of problems, solutions, and forgetfulness. In addition, reforms are also themselves an important source of problems, solutions, and forgetfulness that all breed further reforms. Reforms call attention to problems, either the problems they are meant to solve or to others. Reforms are theories which at the early visionary stage can easily be adapted to popular theories about how organizations should work, and which therefore seem to provide more attractive solutions than current practices. And reforms tend to lose their beauty over time, becoming easy victims of proposals for new reforms. In short, reforms tend to be selfreferential; reforms are important causes and effects of reforms.

THE RISKS OF REFORM Reforms can be perceived as both positive and negative, from the point of view of organizational insiders and outsiders. They can be seen as important ways of producing repentance, hope, and a tolerance of failure in a world where standards are and should be higher than it is possible to achieve in practice. They can also be regarded as futile and expensive attempts to achieve unrealistic goals. Organizational legitimacy and survival may beneﬁt from reforms, from inspiring the hope of a future better than the present (March and Olsen 1983). If customers, ﬁnancers, public authorities, or employees ﬁnd the organization’s present performance unsatisfactory, they may be convinced not to withdraw their support by a reform that promises future improvement. But reforms are not a risk-free strategy from the point of view of

Organizational Reforms as Routines

101

organizational survival. It has been argued here that reforms tend to be easy to start but more difﬁcult to implement, so that they sometimes increase attention to problems and levels of aspiration more than improving behaviour. The main result of the reform is then that it reinforces the perception of failure which was once the reason for launching the reform. Reforms risk increasing the importance of the very goals which the organization has greatest difﬁculty in achieving. While before the reform some might have argued that the value to which the reform was geared was not important or relevant to the organization, after the reform it may become a generally accepted criterion of evaluation. In some cases, this can have a lethal effect on the organization. For instance, an attempt to reform railway companies to become more business-like and proﬁtable might succeed in establishing proﬁtability as the main evaluation criterion but without being able to change behaviour enough so that the organization could fulﬁl this criterion. This last is an unsurprising result, because proﬁtability is the very criterion which railways have found most difﬁcult to meet for the last century. The overall result of reforms aiming at increasing proﬁtability may then be to close the organization down more quickly. In this way, the reform of individual organizations may be a way of changing over to more modern forms, not by changing the organizations undergoing reform but by bringing a quicker death to those which cannot meet modern standards, thus hastening the adoption of modern forms in the total population of organizations. From the point of view of the individual organization there is reason to avoid these effects, if survival is considered more important than the meeting of high standards.

AVOIDING REFORM Organizations thus have reason not only to promote but also to avoid administrative reforms. The argument so far should encourage reformers – it seems to be fairly easy to start reforms. Those wanting to avoid reforms can base their strategies on a variety of threats to reform. For instance, if problems or solutions or forgetfulness are lacking, reform is less likely. But it is difﬁcult to imagine that a lack of problems could be a very common

102

Reform as Routine

obstacle to reforms in large organizations. Rather, one type of problem could even block reform, namely problems that are too acute. Since reforms require a certain length of time, they are more difﬁcult to present as solutions to a threat of bankruptcy today than as solutions to a future threat. Reforms are threatened not only by a lack of problems, but also by alternative ways of handling problems. Reforms imply a special approach to problems. They are aimed at action, change, and problem solving. Problems should be solved by doing something about them – the situation is changed so that problems disappear. Another way of reacting to problems is to ignore them. For example, organizational managements may choose to delegate problems and let other units in the organization grapple with them, instead of intervening with reforms. Problems may also be dealt with by increasing organizational slack (Cyert and March 1963). So instead of trying to solve problems by means of administrative reforms, an organization can try to increase resources, perhaps by turning to more proﬁtable products or improving its marketing. Another way of dealing with problems is to administer them, to live with them. People can analyse them and discuss them without trying to solve them in practice. Ignoring problems is difﬁcult for a weak management unable to resist the demands of subordinate units that it should solve their problems. Administering problems instead of solving them is easier in organizations such as parliaments, which are evaluated more for their ability to provide a discussion of different opinions than for undertaking action (Chapter 4). A shortage of solutions may be perceived by a modern organization which has already introduced the latest structures, processes, and ideologies. It is more difﬁcult for an innovator to ﬁnd new solutions than it is for an imitator. Knowledge of solutions makes reform easier, but too much concrete and complex knowledge of the problems of implementing modern solutions, as well as too much knowledge of their results, may prevent reformers and others from accepting reforms. If reforms can be facilitated by the mechanisms of organizational forgetfulness, they can also be prevented by the mechanisms of organizational memory. Trying to keep an old management in power, avoiding new personnel, and refraining from the use of external people like consultants are all possible strategies for avoiding reform.

Organizational Reforms as Routines

103

If reforms contribute to a tolerance of present failure and the production of hope, they can also expect competition from other methods of achieving these things. One such method is to emphasize the good intentions of an organization or its management, at the expense of descriptions of results. Another method is to emphasize the future rather than the past – to promise improvement (Brunsson 2002). Goals describe both intentions and the future, and they can therefore produce both hope and tolerance. But all these methods may also lead to a demand for reform. Failure to fulﬁl intentions and goals is often blamed on administrative problems, but administrative problems provide arguments for administrative reform. And one way of increasing the credibility of references to a better future is to launch some reforms. If reforms are much more attractive ex ante than ex post, attempts at implementing them will threaten the support they have won. A relatively secure phase in a reform process is at the beginning of its implementation, when it still looks promising. If reforms can stay in the stage at which they have just been started but are still far from ﬁnal implementation, they can survive for a long time. But if serious attempts are made to implement a reform, its opponents can help to get it to a stage where it is clear that the reform solution will not be so beautiful in practice, maybe not even any better than the practice of today. And if support falls off, perhaps the reform can be stopped. The strategy of implementation may halt a speciﬁc reform, but as we have seen, implementation easily leads to a further reform with the same – or the opposite – characteristics as the ﬁrst one. The same effect arises from another important threat to reform – a proposal for another reform. Organizations cannot be expected to pursue several reforms in the same area at the same time. Proposing a reform to stop an existing one is most effective when the existing reform is at a weak stage in its life, when it just has been proposed and not yet attracted too much commitment, or when it has gone far enough on the way to implementation for its less-beautiful ex post characteristics to have revealed themselves. It is then all too likely to be stopped by the proposal of a new reform presented in the more beautiful ex ante version. Promising a new reform may even be the only feasible way of stopping an old one. Important strategies for anyone trying to stop reforms thus consist of providing arguments that easily lead to new reforms, trying to implement

104

Reform as Routine

existing reforms, and proposing new reforms. The opponents of the reform become reformers themselves! Even if it may be possible to stop individual reforms, it is difﬁcult to stop reforming in general. So not only are reforms their own causes and effects, but attempts at avoiding reforms imply support for reforms. Self-reference and opposition make routines of reforms.

6 The Standardization of Organizational Forms as a Cropping-up Process Large modern organizations tend to have fairly elaborate forms, i.e. formal structures, procedures, and ideologies which are presented externally and internally as characteristics of the organization in question. Such forms are not usually unique to one organization but are shared by many. Among organizations, the patterns of homogeneity and heterogeneity in these forms are often complex. Organizations that differ in many other respects may still exhibit very similar forms, while organizations that are essentially similar may exhibit forms that are quite different. There is often little homogeneity over time: many organizations change their forms with considerable frequency. The homogeneity and heterogeneity of organizational structures, procedures, and ideologies can be explained in several ways. Organizations may differ in their forms because they are undertaking different kinds of operations, or because they are working in different environments. But there can also be striking similarities in the forms adopted by organizations with very different kinds of operations or working in seemingly different environments. Similar forms have been observed in organizations involved in quite different kinds of production (Woodward 1965). Organizations in different ﬁelds may exhibit similar forms: for instance, private organizations have adopted many of the forms of public organizations such as budgeting and strategic planning, while public organizations have borrowed from the private sector. And fashions in organizational forms spread rapidly among large companies located in different countries (Mintzberg 1979; Abrahamson 1996).

106

Reform as Routine

These similarities may arise because organizations actually do encounter similar environments and problems, and have found similar ways of handling them (Thompson 1967). For instance, if employees acquire a stronger belief in, and claims to, their own individuality (Thomas et al. 1987), many organizations may invent similar, more decentralized forms as a result. Or if a growing number of people begin to see themselves as customers rather than citizens in relation to their public organizations, then the organizations may start to adopt forms similar to those in private companies. This kind of explanation builds on the assumption that formal structures, procedures, and ideologies are important to the organization’s operations and its ability to solve practical problems. But the relation between organizational form and organizational activities is often weak (Weick 1969; Meyer and Rowan 1977). The ‘decoupling’ of form and operations opens the way to other types of explanation. A key argument in institutional theories of organization is that organizational forms are strongly inﬂuenced by widely held norms and ideas about the kind of organizational forms that are natural, correct, or desirable (Meyer and Rowan 1977; Powell and DiMaggio 1991). Norms and ideas held in common constitute a major impetus to standardization; they promote similarities in many organizations, even those which conduct quite different types of production but which operate within the same culture or the same ﬁeld (DiMaggio and Powell 1983). But the existence of common conceptions is not enough to explain homogeneity in organizational forms. There is no guarantee that all organizations adhering to commonly held conceptions will have similar forms. There is often a combination of homogeneity and heterogeneity within a single culture or organizational ﬁeld: even if some organizations introduce a fashionable form, others do not. So the argument that organizations are exposed to common norms about what makes a good organization cannot fully explain their organizational forms. We need to explain the homogeneity and the heterogeneity that can arise among organizations which do in fact recognize similar ideas. One way of discovering such explanations involves examining the processes whereby general conceptions are transformed into local organizational forms. In this chapter, I will argue that standardization processes are likely to differ when different kinds of ideas and norms obtain. I will pay particular

The Standardization of Organizational Forms

107

attention to what I call informal norms. It is sometimes assumed that such norms give rise to homogeneity via a process of diffusion. I will argue that under some circumstances a diffusion model is not particularly useful for explaining homogeneity. As an alternative, I will present a ‘cropping-up’ model of standardization. In the next section, I will examine various kinds of ideas and norms. I will look at a diffusion model and consider its limitations. In the following section, I will then present the basic concepts underlying a ‘cropping-up’ model: forms, reforms, and organizational discourse. This is followed by a section in which I describe how organizational forms are determined by the interaction between general discourse and local reform cycles. Finally, I will discuss the way local reforms may affect the norms expressed in the general discourse.

STANDARDIZATION PROCESSES Various conceptions about appropriate organizational forms are likely to affect organizations in different ways. Some of the conceptions have a constitutive character, i.e. they are necessary ingredients if an entity is to be regarded as an organization at all, or as a particular type of organization. For instance, it is hard to make people believe that an entity is a real organization if there is no accounting system and no responsible person at the top; likewise, a ‘true’ company must have owners, or an association must have members. Constitutive conceptions are contingent on the construction of organizations. They are relevant only when people consider an entity to be an organization or a special type of organization, but in such considerations these conceptions are a necessary ingredient. Constitutive features are similar in all organizations, or at least in all organizations of a speciﬁc kind, but it still needs to be explained why certain activities or entities come to be regarded as organizations, or as organizations of a different type than before. The existence of constitutive conceptions enables us to explain why a school that comes to be regarded as an organization will acquire a local leadership and a local accounting system, but it does not

108

Reform as Routine

help us to explain why it came to be regarded as an organization in the ﬁrst place. Processes leading to such redeﬁnitions are described in Chapter 3. Other ideas about appropriate organizational forms are expressed in formal rules combined with sanctions. These may be part of the legal system or of professional codes. For instance, organizations of different types in different countries must use speciﬁc forms of accounting. Formal rules create homogeneity through various mechanisms of coercion (DiMaggio and Powell 1983), such as imposition, authorization, or inducement (Scott 1991). Heterogeneity can arise from variations actually stipulated in the rules themselves, or from non-compliance on the part of certain organizations. But there are also standards and norms of a less formal kind, which have much weaker links or even no links at all with national or professional codes and which are combined with weaker or more ambiguous sanctions. Examples include ideas about particular organizational forms that emerge as fashions in the organizational world, such as the matrix organization, total quality management, or service-orientation. Ideas initiated or reinforced by business schools, such as budgeting or human resource management techniques, are further examples. Standards are not a necessary part of the constituting of the organization; nor can they be forced upon it. There is no guarantee that they will always be familiar to local actors who are able to inﬂuence organizational forms, or that these actors will accept them. And yet they affect many organizations and help to create similarities between them. The exact outcome of this inﬂuence depends on the processes involved. It is the process whereby informal norms produce homogeneity and heterogeneity that I shall be discussing in this article.

A diffusion model A possible tool for analysing how informal norms produce homogeneity and heterogeneity is a common one in social research, namely the diffusion model. Diffusion is a standard concept in the social sciences for explaining homogeneity (Strang and Meyer 1993). The concept is often used to describe the end result of a process – the existence of similar forms and practices at different places at a certain point in time. But it is sometimes

The Standardization of Organizational Forms

109

also understood as a description of a process whereby these similarities arise. Something diffuses from a centre to a periphery. Information about new ideas, things, or practices is disseminated to ever more distant places, and this diffusion produces increasing homogeneity among such social entities as have the attributes to render them susceptible to it. The process is somewhat similar to the process of contagion or infection (March 1981): organizations or other entities in close contact with the centre of contagion or with other infected entities are likely to become infected themselves, that is to say they are likely to adopt whatever it is that is being diffused. Unlike an infection, however, the ‘object’ of the diffusion cannot spread by itself – it must be actively adopted by some actor or other (Malinowski 1927; Latour 1986). Knowledge is often assumed to be a crucial scarce resource: before we have come into contact with others, we do not know about the new entity being diffused and cannot therefore adopt it. The diffusion process is thus a process in which actors learn from others. Diffusion processes may well explain how some informal norms about organizational forms are adopted by organizations, thus producing homogeneity in the forms they use. For example, representatives of different organizations may meet one another and learn about each other’s forms, after which they may try to introduce them into their own organizations. Professionals employed in organizations attend professional meetings and can be ‘infected’ in this way. Management and leadership are increasingly coming to be regarded as professional skills, which can mean that managers – those who formally make decisions about organizational forms – meet each other in a similar way. The individual organization may also be infected indirectly by a carrier or transmitter, the equivalent of the rat or mosquito in the transmission of disease. Management consultants, for instance, may spread the same form to one organization after another as they move among them offering their services. Many organizations can be infected at the same time by listening to the same experts at conferences. The basic micro-process whereby organizations adopt the form that is being diffused is sometimes referred to in terms of imitation, i.e. people in one organization try to imitate what is being said about the forms in other organizations, either by representatives of these organizations or by other people (Westney 1987). The imitation may be far from perfect – it could perhaps be described as a translation of the general idea adapted to local

110

Reform as Routine

circumstances (Malinowski 1927; Latour 1986) or as a local edition of the general idea (Sahlin-Andersson 1996). Over time the process of diffusion produces increasing homogeneity. But the model also explains heterogeneity among organizations. Like the similarities, differences among organizations are dependent on time, distance, and susceptibility. During the diffusion process, there are differences between the organizations: some, which have not (yet) been infected differ from those which have. Organizations with a limited contact network or located in the periphery are less likely to be infected early on or even at all. Some differences may persist because certain organizations are immune – perhaps they have internal characteristics which make them less apt to adopt a speciﬁc form, such as not being competent, rich, or willing enough. They may be conservative and resistant to change. Or – if there is some connection between forms and operations – they may have a type of operation and a local environment that do not ﬁt the new form. The diffusion model is based on a physical metaphor. Contact with speciﬁc others who possess some particular knowledge is a key explanatory element. Traditionally, the model has been used in explaining fairly slow processes whereby new practices in agriculture, for instance, spread to distant places and become dominant there (Smith et al. 1927).

Limited applicability The diffusion model described above may provide a good way of describing certain phenomena. But under modern conditions its applicability seems to be somewhat restricted: information on many issues spreads almost immediately and there are often an enormous number of entities which deﬁne themselves in the same way and therefore easily adopt new ideas said to concern their type of entity (Strang and Meyer 1993). For instance, knowledge of organizational forms is not a very scarce resource in modern societies: ideas and information about organizational forms are often spread almost instantly and worldwide to an extremely large number of entities perceived as organizations. These are all ‘infected’ at almost the same time. And yet by no means all of them adopt the new form, and the group of adopters may appear to have little else in common. Adoption

The Standardization of Organizational Forms

111

may take place at about the same time almost everywhere, so adopting organizations have little chance of imitating each other. More speciﬁcally, the diffusion model is less useful for explaining the kind of homogeneity that arises without any contact between similar organizations or contact with a common transmitter, or when knowledge is not a scarce resource. It is less useful when widespread knowledge of a certain form, or even its popularity, does not result in much homogeneity, i.e. when a form that is popular and fashionable in the public debate is not actually installed in many local organizations. The diffusion model is also less useful when the micro-process is one not of imitation but of innovation, i.e. when individual organizations introduce forms which no other organization has yet tried, or they invent a new form, and still they end up with forms that are similar to those of other organizations. Nor is the diffusion model very useful when organizations appear to lack any stable quality that could explain whether or not they are susceptible, for instance when organizations have very similar operations and local environments and still exhibit different forms, or when a certain organization resists a popular form on one occasion and on another is quick to adopt some other very similar form. An empirical case in which the diffusion model seems to be of little use is provided by the changing forms that have been adopted by Swedish municipalities over the last few decades. In the public debate, new ideas about organizational forms have popped up at intervals of a few years (Johansson and Johnsson 1994). The public debate was familiar to municipal managers (Fernler 1996) and it can be assumed that they learnt about the new ideas at about the same time. A new form was often adopted by many municipalities during the short period when the form in question was popular in the public debate (Johansson and Johnsson 1994); most adopting organizations had no other organization to imitate. However, most forms were taken up by a minority of the municipalities only: the extent to which municipalities adopted any one speciﬁc fashionable form varied between 6 and 70 per cent (Johansson and Johnsson 1995). Further, it was difﬁcult to ﬁnd any strong correlation between the forms adopted, or any speciﬁc qualities of the municipalities concerned such as size or the identity of the political party in power (ibid.). Some of the new forms arose at about the same time in other countries, in the public debate as well as in municipalities there (Hood 1991; Reichard

112

Reform as Routine

1995). Some forms seem to have been invented in several municipalities at about the same time (Fernler 1996). In order to explain such processes we need a model which differs in several respects from the diffusion model. The one we require should be able to explain homogeneity arising through processes other than imitation among organizations which lack contact with one another or a common transmitter. It should be possible to use the model for explaining not only homogeneity but also heterogeneity, i.e. why some organizations do not adopt a particular common form. All in all the model should explain something that we could call ‘cropping up’ as opposed to diffusion, that is to say the scattered and seemingly random appearance of similar forms in many but far from all organizations at the same time. In the next section, I will describe one version of such a cropping-up model.

A CROPPING-UP MODEL OF ORGANIZATIONAL FORMS The model to be outlined below contains three basic components: organizational forms, reforms, and discourse. The model contains speciﬁc assumptions on the nature and dynamics of these components. It is based on the assumption that, fundamentally, organizational forms are not things or practices, like the items often described as spreading in the diffusion model; rather, they are representations. They must therefore follow the rules regarding the way we can talk about organizations. The reform processes whereby these forms are introduced require even closer adaptation to the rules. They also require attempts at implementing the talk in practice – attempts which are often time-consuming and frustrating. The common component that makes homogeneity possible is assumed to be a broad societal discourse on organizations. Here some forms are more popular or fashionable than others. But popularity or fashion does not necessarily lead to homogeneity. Very popular fashions in the discourse may not be reﬂected in a high degree of homogeneity in organizational forms (just as a haute couture fashion may not be reﬂected

The Standardization of Organizational Forms

113

in many people’s way of dressing). The model will describe the mechanisms whereby popular forms are, or are not, translated into local forms.

Organizational forms Organizational forms are part of the way organizations are represented to the external world or to their own members by authorized people such as managers. Organizational structures are often displayed in organization charts, while procedures appear in documents containing rules and ideology in established and documented policies. These written sources are normally conﬁrmed by oral presentations by management and by labels denoting entities, practices, or ideas. The forms are presented as descriptions of the way the organization actually works, or the way it will be working in the near future. The forms also tend to be highly rationalized, being typically presented as means to legitimate ends or as solutions to important problems. Or at least rationalizations are readily available upon request (Meyer and Rowan 1977). The extent to which forms describe actual practices and the extent to which practices and forms affect each other, will differ from one form, organization, or practice to another; they will also differ over time. As noted in the introduction above, forms and operations are often ‘decoupled’ and are not very consistent. Practice tends to be less affected by the form than the presentations suggest, or to be affected differently. And the efﬁcacy of the forms in achieving the ends and solutions presented tends to be weaker, rarer, and more uncertain than described in standard rationalizations (March and Olsen 1989, Ch 5). So if we adopt the basic distinction between organizational talk and action (Brunsson 2002), then we can assign organizational forms to ‘talk’. Just as there are norms and rules for action that restrict our freedom to act, so there are norms for talk that restrict the way we can talk, at least if we are to be believed or taken seriously (Brunsson 2007, Ch 8). In the model it is assumed that this is true of talk about organizations, and thus of organizational forms. Organizations cannot then be represented just anyhow: the representation must be adapted to ideas about what an organization can and should be. The forms must be such that they can be regarded as possible organizational forms; they should be adapted to

114

Reform as Routine

what people consider good or at least acceptable. In addition, they should be understandable: excessively complex or ambiguous forms are not likely to be understood.

Reforms The processes whereby new forms are launched will be referred to below as reforms. Reforms as deﬁned here consist of two basic activities: attempts at convincing people that the new forms should be installed and attempts at getting them implemented. Studies of many such reforms in a variety of organizations (Brunsson and Olsen 1997) revealed a certain pattern (partly described in Chapter 5), which will be used here in formulating some assumptions about the reform component in the model. The extent and intensity of talk differed as between existing forms and new ones. Reforms required much more mobilization and talk than existing forms. Typically, reforms produced a lot of elaborate talk about the forms-to-be, and a good deal of rationalizing argument. Reforms were well adapted to general, aesthetic norms of talk, such as those referring to logic, consistency, and clarity. Reforms were also adapted to what was generally considered to be a good organization. They promised major improvements in the way the organization functioned. At the outset, the reforms were described as a set of simple, general, and seemingly very sensible not to say beautiful principles; for instance, the organizations were to become more goal-oriented, everybody’s tasks should be clearer and should not overlap with the tasks of others, everyone’s freedom to act efﬁciently should be increased, better planning and evaluation processes should be installed, etc. The principles contained elements which were popular and not controversial at the time of their introduction, such as decentralization or management by objectives. The situation before the implementation of the reform was described in a less favourable way: it was said to be complicated, tasks were unclear, there were inefﬁciencies, lack of local freedom, etc. In sum, it seemed that reforms not only produced more talk than existing forms, but they also followed the norms for talk more closely. This can be explained by the fact that reforms are future-oriented and intentional. The way we can talk about existing forms may be limited to

The Standardization of Organizational Forms

115

some extent not only by the rules for talk, but also by practice. It may be necessary to adapt talk somewhat to what people actually do, in particular if the talk is directed at the doers themselves. Reforms, on the other hand, can be more easily adapted to the norms geared exclusively for talk, since they are not meant to reﬂect current practices. Reforms include descriptions of practice: their implementation and their effect on practice is the standard argument in their favour. But this practice is still in the future, and can thus easily be adapted to the new forms instead of the other way around. But reforms not only can but must be well adapted to norms for talk about organizations. Reforms contain intentions; the forms presented are those we intend, and they are to lead to the effects we intend. The norms regarding the intentions which we can present are generally stricter than those applying to our representations; for instance, we may admit to bad results but we are less likely to say that we intended them. This may be true of reforms as well: we may represent current organizational forms as less than perfect, but it is more difﬁcult not to intend to be as near to perfect as possible. It is not being claimed that all empirical reforms are like those described here, but the model includes this type of reform. It will be assumed that organizations are attracted by such reforms. Since these reforms can be much more beautiful than possible representations of today’s practice, most reform proposals will prevail over the defence of most old forms. But it is still assumed that there is one major obstacle to reforms, namely previous reforms. Reforms are supposed to be implemented, i.e. the principles of the reform should be turned into detailed instructions and should affect organizational behaviour, and even results; and this takes time. It is hard to argue for a new reform until the previous one has been given a serious try, particularly if the reforms concern the same activities. However, this obstacle is only temporary. When the new forms become existing forms, they have to be adapted somewhat to current practices, and they are therefore likely to become less attractive: a new form will easily appear more attractive. And a new reform does not necessarily have to wait for the full implementation – or, even less, for the results – of the old reform; the old reform can become relatively unattractive at a much earlier stage. This is because implementation is assumed to be an important threat to reforms. When the beautiful principles are turned into more

116

Reform as Routine

detailed instructions and are adapted to the special practical conditions of the speciﬁc organization, the new forms soon become less beautiful and more like the old ones, thus losing much of their attraction. Their promises do not seem to be being fulﬁlled. The new forms may also become more controversial. Further, principles adapted to norms for talk are not necessarily feasible or favourable in practice. This may become obvious long before the reform is regarded as fully implemented – or, even more, before it has given rise to any intended effects on daily practice. So, after a serious attempt at implementation, the reform may fall into disrepute and may well be interrupted by a new reform (Chapter 5). When reform processes look as described here, and when there is no lack of new reform ideas, a speciﬁc pattern of reform can be established. A reform starts when the previous reform has been in the process of implementation long enough, and it ends when its own implementation has been under way for a while. The organization is almost continuously in the throes of some stage of a reform, but reforms can only be launched at intervals. The assumption of implementation as a threat to reform and the resulting pattern of reform also reﬂected the empirical situation in the reforms studied. Reforms easily prevailed over older forms, often over those still being implemented. A common way of ending a reform was to start a new one. In some large organizations, reforms were a standard recurring activity. Major reforms occurred often but not constantly – generally at intervals of a few years. A study of the 125-year history of one large organization showed that major reorganizations took place at intervals of 3–15 years, and that the intervals had been shrinking more recently (Brunsson et al. 1989). Although it is assumed that reforms easily win over existing forms, a speciﬁc reform proposal may meet competition from other reform proposals which also contain attractive principles. It is assumed that such competition will be won by reforms which ﬁt better at a more detailed level with the norms for talk about intended forms, and for which more favourable and less critical arguments are available. Such norms and arguments are largely produced and reproduced in an arena other than the individual organization, namely in a general public discourse.

The Standardization of Organizational Forms

117

Organizational discourse It is not only people in individual organizations who talk about the present and future forms in their own case; there is also a more general discourse in modern society about these things. Organizational forms in general are a popular topic in higher education where it touches on organizations, as well as among the relevant commentators – not least in business magazines and management books – and among those involved in helping organizations to reform, such as management consultants (Alvarez 1991). Although the discourse varies somewhat from one organizational ﬁeld or country to another, much of it is very general and treats large groups of organizations as though they were in need of the same forms (Furusten 1995). The discourse is to a large extent international, i.e. national borders are not thought to be important and much of the discussion takes place across such borders. Organizational problems and solutions are both discussed. The discourse produces ideas and norms about possible and desirable organizational forms. Although the forms discussed are likely to be based on some common basic conceptions of organizations, they differ when it comes to more detailed aspects. For example, all the forms discussed may be based on the assumption that organizations should be controlled from the top, but they may still differ greatly about the form that is best for bringing this about. In this discourse only a limited set of ideas can be attended to at one time. The ideas attended to can be variously controversial, and the level of controversy may change over time. Ideas can also differ in their level of abstraction. Very speciﬁc organizational forms are sometimes suggested as the ‘right’ ones. Zero-based budgeting might be an example of this. In other instances, the suggested forms are of a more principled and abstract character, more in the nature of slogans than recipes. ‘Back to basics’ – concentrating on the ‘core’ business – and decentralization are cases in point. At an even higher level of abstraction, the discourse may concern problems but no speciﬁc solutions: there may be a large measure of agreement in identifying the important and urgent problems, without any suggestion being made about speciﬁc forms that could perhaps provide a solution. The attention paid to problems varies just as much as the attention paid to solutions. In the 1970s the lack of internal democracy in organizations was a popular problem: later, inadequate

118

Reform as Routine

customer orientation and cost-efﬁciency were among the problems in vogue. Knowledge of this general discourse is not ‘outside’ the organizations; people working on the managing and reforming of organizations are aware of the discourse and may even take part in it. In the cropping-up model, unlike the diffusion model, it is not assumed that knowledge is a scarce resource. Rather, it is assumed that people interested in organizational issues will hear very quickly about any new arguments, regardless of where these originated: they will all be reported widely in the media. This kind of discourse is thus added to our model. We assume that a set of organizations is connected to a common discourse. The discourse provides norms and arguments regarding what are acceptable or good organizational forms, about forms that should be striven for and problems that require solutions. In addition, people participating in the discourse may have an authority that can be referred to in arguing for reforms in the right direction. The norms and arguments in the discourse help to explain which reforms are chosen by individual organizations: they may serve as an important inspiration for reformers who are designing a new reform, and they can determine which of several reform proposals is accepted by an organization. It is assumed that the discourse evolves continually. New ideas are attended to, others become uninteresting. A basic mechanism behind changes in attention may be a focus on novelty: when no further descriptions and arguments can be added to the discourse, there is not much left to discuss and attention can be directed towards a new idea. Some ideas move from commanding general agreement to becoming highly controversial, others move in the opposite direction or retain a fairly stable element of controversy. Finally, some ideas even become generally regarded as bad. The discourse can move towards greater speciﬁcation, i.e. abstract principles are given a more detailed, concrete content; problems are given solutions, or concrete cases are quoted in which principles are said to have been transformed into practice. Or the discourse may move towards more conﬂict, as counter-arguments are proposed. These two tendencies may be interrelated as they were in the reform processes in individual organizations described above, so that more speciﬁcation produces more controversy. Speciﬁcation and controversy may be replaced by new abstract ideas or by compromise, and perhaps by abstraction again.

The Standardization of Organizational Forms

119

MODEL DYNAMICS The cropping-up model is thus based on speciﬁc assumptions about organizational forms, local reforms, and a general discourse. Organizational forms are ways of talking about organizations. Reforms, the presentation of new forms and attempts at implementing them, have to be adapted to the norms for talking about organizations, and particularly to talking about organizational intentions. Some of these norms are provided by a general and continually evolving discourse about organizational forms. But due to their connection with talk and discourse, reforms are apt to lead to disappointment when it comes to implementation: what is considered good talk does not necessarily correspond to good or even possible local practice. Moreover, what is good generally may not be good for the individual organization. The result of all this is recurring reforms in a continually evolving discourse. There is thus a certain dynamic. At any one moment we have a set of organizations ripe for reform, i.e. organizations which have gone so far in implementing previously popular forms as to create frustration and an interest in new forms. At every such moment all these organizations are exposed to the same discourse, with its speciﬁc set of more or less controversial and more or less speciﬁc ideas. From the discourse it is possible to gather ideas about forms currently attracting attention there, and to ﬁnd arguments and support for them. Reform proposals implying fairly uncontroversial forms will have the advantage over reform proposals along other lines. The level of attention and of controversy affects not only the organizations’ choice of forms, but also to some extent the area – organizational structure, perhaps, or budget procedure – in which they reform, since not all areas are necessarily being attended to or are sparking uncontroversial ideas at every point in time. Since the discourse is continually evolving – creating new ideas, new controversies, and new speciﬁcations – the set of organizations which go in for reform during the next period, will not introduce the same set of forms. Processes such as these could explain both homogeneity and heterogeneity between organizations. The explanation refers to no attributes of the individual organization apart from the timing of its reform cycles.

120

Reform as Routine

Organizational reforms are similar in organizations reforming at the same time, and different in organizations reformed at different times. If the implementation process differs somewhat in length in different organizations, perhaps due to some random external variables, organizations which have embarked on a certain reform simultaneously will not be ripe for a new reform at quite the same time. It is then slightly less likely that these organizations will choose the same reform as the others in the second round. This makes for variation over time as regards the organizations which any speciﬁc organization is likely to resemble. In other words, it impedes any order among organizations on a cohort basis, and conﬁrms a picture of great variability and randomness. The model explains how popular ideas in the general discourse are translated into local organizational forms. It also demonstrates why popular forms are not necessarily very common in organizations: they are adopted only by organizations which happen to be ripe for reform during the period when the forms are popular.

Processes producing homogeneity The processes whereby forms are transferred from the general discourse into the local organization should also be speciﬁed. One of these processes can be called adoption. In diffusion models (and in my description of such models above), the term ‘adoption’ generally refers to the outcome of a process. In our present model, instead, adoption describes a speciﬁc process whereby the local organization takes over a form described in the discourse. Another possible process involves invention, whereby a form is constructed locally. If the uncontroversial ideas in the discourse are fairly abstract and do not deﬁne any particular forms, the individual organization has to invent its own. But even if the ideas are abstract, they may contain problems and ideas about the present situation or about organizations in general which are speciﬁc enough to make most reformers in most organizations come to very similar conclusions regarding what form to introduce. The forms are invented locally, but are still very similar. There is a parallel here with certain inventions and discoveries in science: oxygen, for example, was discovered by at least two scientists simultaneously (Kuhn 1962). Similarities arising from inventions are

The Standardization of Organizational Forms

121

particularly likely when there is a strong consensus about what problems are important, but no agreement on speciﬁc solutions. Organization members may well see their own invented form as something unique at ﬁrst, only to learn later that it is in fact quite common. In this case they may perceive themselves as the ﬁrst to introduce it, and perhaps develop a theory about how they have inﬂuenced other organizations. When such forms enter the general discourse they have to be generalized to some extent. Talk of a more detailed kind in the individual local organizations may vary somewhat, but can still lend itself to a common description at the discourse level. In the adoption case, reformers advocate the same relatively speciﬁc forms as those attended to in the discourse. Even though the forms here are more speciﬁc than in the invention case, they are seldom so speciﬁc that they cannot be interpreted a little differently and adapted to local conditions and taste, just as in the case of diffusion. Again this allows some scope for slightly different local versions, without disturbing the impression that the local reform ideas are in fact the same as those being generally discussed. The adoption process can follow different routes. It may be a question of inspiration, whereby local reformers gather their ideas from the general discourse. Or it may be a question of labelling (Czarniawska-Joerges 1988), whereby reformers adapt their ideas (regardless of their source), as well as their talk about their ideas, to the forms currently appearing in the general discourse. Inspiration often involves a process of speciﬁcation in which general forms are translated into a local version; labelling normally involves a process of generalization, in which speciﬁc local ideas are given more general labels.

Frequency of forms Given the processes described above, the number of organizations using a particular form will depend mainly on the length of time during which a particular form is comparatively uncontroversial in the discourse. The longer a form or a problem is considered to be the right one in the discourse or the greater the frequency of its being so considered, the greater will be the number of organizations which will have embarked

122

Reform as Routine

upon a reform period during this particular discourse, and the greater the number which will have adopted the form or invented similar forms. In this way (length of) time translates into (width of) space. The frequency of a particular form also depends on whether or not competing forms appear. If new forms do compete, the old ones are ultimately likely to be erased. For instance, an old principle for structuring the whole organization cannot normally be retained when a new such principle is installed. It is hard to be basically centralized and basically decentralized. When the competitive aspect is weaker, an old form can be retained together with new ones, so long as it does not fall into disrepute for other reasons. A form becomes very common if it is attractive and remains uncontroversial for a long time, and if competitive ideas do not appear for a considerable time after the attention giving rise to the reform has faded away. Budgeting can provide a relevant example here. It became popular for use in the private sector in the 1960s, and generated a great deal of literature over a long period. It has proved possible to combine budgeting with various other organizational forms, including other management accounting procedures, and it has not so far fallen into serious disrepute. Thus most organizations still present budgets, although budgeting is no longer receiving much attention in the discourse. Budgeting has even become highly institutionalized, almost a constitutive norm; having formed the basis for special departments etc. in many organizations it is now taken for granted and could not be easily questioned in the discourse – or such questioning would not affect organizations very much anyway (Wallander 1994). Similarly, organizations may maintain that they hold ‘development talks’ with their employees long after everybody except a few specialists have ceased to refer to the technique as important. In this way, forms which may once have appeared quite suddenly can become sedimental, continuing to exist although they are no longer paid much attention (Danielsson 1983).

Cropping-up explained The cropping-up model is not intended for use in explaining all homogeneity or heterogeneity in organizational forms. But it can sometimes

The Standardization of Organizational Forms

123

help to clarify certain phenomena which the diffusion model, or the model mentioned in the introduction which relates operations, environments, and forms, cannot altogether explain. The cropping-up model provides a possible or partial explanation of the more or less simultaneous appearance of similar forms among organizations which have no contact with each other and which lack a common transmitter. And when we use this model for explaining homogeneity and heterogeneity, we do not need to know any speciﬁc characteristics of any individual organizations apart from their position in the reform cycle. Further, the model can explain instances of standardization arising from processes other than imitation.

INTERACTION BETWEEN REFORMS AND DISCOURSE In the cropping-up model it is assumed that the general discourse affects local reforms. But in reality inﬂuence may also work in the opposite direction. Local reforms may inﬂuence organizational discourse; experience of local reforms may be generally discussed. Local organizations may in fact reinforce the popularity of certain forms by reporting their own reforms and proclaiming their success. But local reforms in line with the current discourse may also introduce controversy about a particular form, or even engender agreement on its negative aspects. The form may then be abandoned, leaving room for a new idea. As in the individual organizations, attempts at implementing reform may make general principles seem less attractive. There may be several reasons for this. First, reforms can lead the discourse towards a higher level of speciﬁcation. As has been noted, local reform generally calls for a more speciﬁed version of a general form, or a general problem must be met by a speciﬁc form. The speciﬁcations may vary from one organization to another; if these different speciﬁcations appear in the general discourse, there will be a higher risk of controversy than before. Different people may advocate different speciﬁcations, which can reduce conﬁdence in any speciﬁcations. But this process depends on the way the speciﬁcations are reported back in the general discourse. The more general the terms in which they are

124

Reform as Routine

described, the less impact they have on the discourse. If the discourse is dominated by top managers possessing little information about details or consultants or researchers skilled in generalization, then controversy is less likely than if the discourse is joined by people who possess more detailed information and who are less prone to generalize. Secondly, information about local attempts at implementing reform can be expected to produce a more mixed discourse on implementation and results. Without empirical evidence, agreement on any particular form is based on the expectation of its relatively easy implementation and its positive results. If empirical observations are included in the discussions, there is a greater risk that information about difﬁculties and negative results will emerge, thus producing a less favourable picture which may then increase controversy and produce many arguments against that particular form. Difﬁculties in implementation can be expected to be more threatening than negative results. Assessments of results tend to be more ambiguous, and hence less obviously negative. On the other hand, in periods of general decline and difﬁculty, ambiguity about the results of reform may turn reforms into scapegoats, thereby producing agreement on how ‘bad’ they are and making room for new ideas. Thirdly, reforms in the same direction in a large number of organizations may change perceptions of the form in question among organizational actors and in the discourse. When the news spreads that this form is being implemented in many organizations of different kinds, the invention process is no longer possible, and the ways in which the form can be introduced become limited. But, what is more, the form may now come to be perceived in some cases as a ‘fashion’ rather than a rational solution; its adoption may be criticized as an instance of imitation, which is difﬁcult to combine with the standard conception of the organization and its management as rational problem-solvers (Abrahamson 1996). Fashions are also difﬁcult to combine with the basic idea that organizations have strong identities of their own, with particular or even unique characteristics and tasks. Such a view of an organizational form involves negative arguments in the discourse, making it more difﬁcult to implement that form locally. Fashions are more powerful when actors do not perceive them as fashions (Røvik 1996). These mechanisms do not always work: in some cases the high frequency of an organizational form has the opposite effect of reinforcing

The Standardization of Organizational Forms

125

agreement that it is a good one. On the other hand, it is not likely that attention will be paid to such a form in the discourse for much longer. Fourthly, the discourse may be affected by an active local demand for new forms. After a while there will be a number of organizations which have failed in the implementing of a speciﬁc reform, or which have recognized the absence of results and which therefore enter upon a phase of new reform. To borrow a term from the diffusion model, we could say that previous reforms may make organizations immune to the form in question. It will be difﬁcult for reformers to produce the same reform again, so they will look for another area of agreement in order to be able to launch a new reform. In most of these cases, local reforms create, or help to create, changes in the content of the general discourse. This in turn provides a basis for new reforms, which in turn tend to change the discourse again, and so on. In most of the circumstances described above, the system as a whole becomes very dynamic and highly diversiﬁed: the discourse changes at a great rate and individual forms are therefore chosen only by a few organizations, which in turn gives rise to diversity.

SOME ADDITIONAL QUESTIONS In this chapter I have mentioned three ways of explaining homogeneity and heterogeneity in organizational forms: by referring to operations and environments; to diffusion processes; and to cropping-up processes. The ﬁrst explanation builds on the assumption that forms and operations are closely connected, the other two on the assumption that forms and operations may be fairly independent of each other. Some homogeneity and heterogeneity among organizations can be explained by versions of one of these models only. Other cases can only be explained with the help of more than one of the models. Detailed empirical studies of reform processes can give us insights into the way the conditions and processes described in the models interact in practice. All three models explain homogeneity by referring to a common element among the organizations concerned. They work under similar operational conditions, they have been infected by the same idea, or they share a common discourse. But the models do not explain much about the way

126

Reform as Routine

these common underlying elements arise or develop; rather, they indicate that this is an urgent topic for research. In the ﬁrst model, it is important to investigate how similar conditions come about. In the diffusion model it is important to explain why something ‘diffuses’ at all, and why it affects a particular population of social entities. In the cropping-up model, it becomes important to explain how the organizational discourse evolves over time and how its perceived relevance for various organizations develops. All these questions call for empirical studies at a higher level than that of the individual organization.

7 Reform and Power The essential idea behind organizational reform is that reformers are powerful – that they can freely choose and create new forms that improve operations and lead to better results. In this chapter, I draw upon observations in previous chapters to examine the extent to which this idea is a realistic one. I begin by examining the assumption of the reformers’ power over organizational forms and operations, discussing the causes of reforms, their contents, and their consequences. In a second section, I consider the role of reforms in image building, and question the assumption that reforms are primarily a matter of change.

REFORMERS AS AG ENTS If power is deﬁned as the capacity of individuals to transform their intentions into practical actions, reformers seem to be powerful people. If they are to be truly powerful, however, reformers must be able to initiate or prevent reforms, to determine their content, and to ensure that they are implemented and that they produce the intended effects. But as the previous chapters indicate, the position of reformers does not tend to be strong enough for them to accomplish all this.

Power over initiation As elaborated upon in Chapter 5, reforms are a routine activity in many large organizations. Reforms could even be said to constitute an integral

128

Reform as Routine

part of the institution of Organization in its modern version. Reforms are expected in the setting of organizations, and it is assumed that major changes in forms and operations will be directed and controlled by managers. Large organizations often include a special department with a mandate to consider, propose, and implement reforms. Organizations that sell reforms – management consultancies – belong to a large and expanding ﬁeld. There are many powerful incentives to reform in the shape of problems, solutions, and forgetfulness. Because reforms do not necessarily lead to satisfaction, they do not pre-empt the need for future reforms. On the contrary, the implementation of a reform may provide a strong incentive for new reforms. Seldom does it generate a state of equilibrium. Contemporary organizations are conspicuously open. Most large organizations publicly display their formal structures, processes, and ideologies, making them available to the scrutiny of many people inside and outside their boundaries – people interested in these forms, and whose positions make their opinions crucial for the success of the organization (Power 1997). The organization encounters external norms and standards for acceptable organizational structures, processes, and ideologies, and it has strong incentive to adapt to them. Its very openness facilitates such adaptation; it is easy to see which forms are legitimate, and the openness of other organizations encourages imitation. When external standards change or when an organization is obviously unable to match prevailing standards, the pressure for reform becomes almost irresistible. The routine nature of reforms, together with their environmental ties, leaves managers with little choice but to accept reforms. The initiative of reformers does not appear to be the main reason why reforms occur; there must be people to drive them, but the fact that reforms take place has little to do with these people. Either they are interchangeable (if they had not pushed the reform, someone else would have done so) or there are norms that compel them to push the reform (a management that always refuses to reform poses a serious threat to its own standing and that of its organization). Reforms are driven by a social setting that is wider than the individual organization, and they are driven by previous reforms. The fact that reforms do occur is often an expression of the powerlessness of an organization, therefore, rather than the power of the organization, its management, or its reformers.

Reform and Power

129

Expressed another way, little personal power is required to initiate reforms. Keeping reform off the agenda, on the other hand, calls for a concentration of power that rarely exists in contemporary organizations. Studies of accounting systems provide examples. In the last century, the design of municipal accounting in Sweden had been decentralized to the municipalities and was the subject of a great number of reforms. In Norway, on the other hand, it had been controlled by a strong state that was able to avert reforms – except in the 1940s when the Norwegian state was taken over by the German state, which reformed municipal accounting according to its own system (Olson 1997). Comparing the SKF corporation up to the 1980s with the accounting systems developed in Swedish municipalities reveals a similar difference: an organization with high power concentration and a virtual absence of accounting reform on the one hand and organizations with little power concentration and hectic reforming on the other (Bergeva¨rn 1989).

Power over content If reformers are to have power, they must enjoy a certain independence in their intentions; they must determine the content of the reforms they promote. In previous chapters, I have indicated a number of factors that reduce the freedom of reformers to do so. General notions about the possible and desirable shape of organizations largely determine what the reformers wish to accomplish. People in positions that allow them to be reformers may hold such notions, but irrespective of their wishes, reformers are generally forced to formulate the content of their reforms to make them acceptable to others, including the reformees. Reforms are part of an organization’s image, of the organizational representation of itself. So reforms must be presentable; they must agree with what is considered good organization. As argued in Chapter 1, reformers who recognize the value of ambiguity and complexity in their organizations would ﬁnd it difﬁcult to present a reform with this thrust. If their reforms are to be accepted, reformers are usually obliged to argue in favour of clarity and simplicity. Moreover, it is difﬁcult to drive reforms in any direction that is not in accordance with the institution of the Organization. It may also be difﬁcult to go against

130

Reform as Routine

the fashion of the times. And fashions comprise both problems and solutions. When quality is fashionable, for instance, it is difﬁcult to argue that one’s organization has no quality problems, and it is difﬁcult to abstain from reforming towards improved quality using fashionable standards for that purpose. All in all, the abundance of ready-made problems and ready-made solutions to the problem of designing organizations makes it difﬁcult for reformers to choose new solutions of their own. Ready-made problems and solutions make it easy to initiate reforms, but difﬁcult to control their content. Reformers may feel forced to adapt the contents of their reform to whatever is feasible if they want to achieve implementation and an image of success. If the people at whom the reform is aimed are also allowed to participate in formulating its content, the chances of implementation increase (Lien and Fremstad 1989), but this procedure substantially limits the power of the reformers. The reform with the greatest chance of success is the one that involves a development which would have occurred in any case or even one that has already occurred. The ambition to succeed can therefore result in reforms that simply describe and interpret organizational change, rather than controlling it. Rombach (1997b) described a reform – the Active County Reform – in which the reformers actively refrained from shaping content, merely describing a change project that was to obtain its content during the process. This reform was considered highly successful by evaluators both inside and outside the organization. The power of reformers is greater when selecting a speciﬁc reform recipe. The reformers’ discretion is highest if several competing standards are offered, without any one of them being clearly more fashionable than the others. Chapter 4 demonstrated how reformers are able to combine elements of various organizational subinstitutions, mixing elements from the State, the Firm, and the Association. Organizations can change their institutional afﬁliation almost completely, but there is every indication that this is normally a slow and intractable process that cannot be achieved through isolated reforms. For the Swedish savings banks to be transformed into full-ﬂedged commercial banks, for instance, took several decades (Forssell 1992). Even in an organization like Swedish Rail, which was originally a combination of state authority and business enterprise, attempts at redeﬁning the

Reform and Power

131

organization in the direction of one single institutional afﬁliation – as a pure business enterprise – met with considerable difﬁculty and took many decades to effect (Brunsson et al. 1989).

Power over effects In the cases that form the basis of this book, it did prove possible to change some organizational forms, although it was sometimes a slow business. Organizational charts were redesigned, budget processes were altered, and new goals and ideas proclaimed. But it turned out to be much more difﬁcult to use the new forms to inﬂuence operations in the desired direction. And it can prove more difﬁcult yet to affect organizational results by introducing reforms. These observations are far from unique. The literature on reform implementation is rife with examples of implementation problems (Winter 2003). The fundamental problem of achieving intended effects is that the rules for what we want and say are often different from the rules for what we do. Because the contents of reforms are decided by the ﬁrst two sets of rules, there is a high risk of failure when these contents meet the world of practice. What is perceived as an ideal or an attractive standard in the world of ideas may not work in practice. Clarity and simplicity are perceived as being key factors in organizational intentions and representations, for example, but they are not necessarily realistic or beneﬁcial in organizational practice. Furthermore, standards forming the contents of reform tend to be general. They may be realistic and beneﬁcial for some organizations but unrealistic or detrimental for many others, because there is greater variation in practice than there is in ideas. Received truths about organizations tend to be extremely general – are thought to be applicable in almost all organizations (Holmblad Brunsson 2007). But organizational operations are much more specialized and differentiated than these general ideas would suggest. Large modern organizations distinguish well between the formal and the informal, and are good at producing ritual and hypocrisy (Meyer and Rowan 1977; Brunsson 2002; 2007, Ch 7). These qualities protect the operations of the organization from many of the demands imposed from above and outside the organization, and they protect operations

132

Reform as Routine

from being affected by reforms. They create ‘implementation problems’ – problems that can, in fact, be seen as solutions. By refusing to implement ideas that do not ﬁt local practice and would not improve it, organizations can keep their operations relatively undisturbed. Organizations with members who are disobedient and ‘resist change’ (Drazin et al. 2004) in this type of reform are better off than are organizations with members who have the opposite tendencies. Implementation problems may ﬁll reformers with despair. But they can also offer some consolation. The very fact that organizational operations are difﬁcult to change by reform means that organizations are resistant to the mistakes of reformers. Organizational reforms are unlikely to be the primary cause of an organization’s demise. Moreover, failed reforms make it easier to start reforming all over again. Reformers are not always interested in discovering the effects of their reforms on operations, perhaps because of the difﬁculty of the endeavour. The important effects are expected to materialize over a relatively long term, and when that time has come, it is difﬁcult to separate the effects of the reform from the effects of all other events that have occurred. Moreover, the reformers’ attention has shifted to other issues, like the one or several new reforms that were initiated in the meantime. Furthermore, the reformers’ objectives may have changed, so they are unwilling to demonstrate the effects of the reform, even if they have been successful according to their initial objectives. In the Swedish state, there has been a long tradition to call for ‘evaluations’ of major reforms to be conducted at a very early stage of reform implementation. This may seem like an illogical strategy, and it poses serious problems for the evaluators, who have virtually no effects to evaluate, but perhaps it expresses the awareness that ‘effects’ are most important before they have occurred. Effects may be part of the reform, rather than something to be investigated, as exempliﬁed by the Active County case mentioned previously in this chapter. When the reformers described the reform, it was not only to mention its aims and the new forms it had generated, but also to talk about its effects. ‘Effects’ occurred before and during the reform, and could therefore be used as internal and external propaganda for the reform; they were not to be investigated after the reform had been implemented.

Reform and Power

133

Institutions not only embrace certain ideas about values, forms, and behaviour, they also hold strong ideas about effects. Efﬁciency, for example, is sometimes regarded as an integral part of the institution of the Organization – and even more of the Firm – rather than a result that has to be demonstrated empirically. Efﬁciency is simply expected to materialize as soon as the organization has become a ‘true’ organization or a true ﬁrm through reform. Then there is little pressure to bother about and demonstrate the effects of the reform. The reforms described as New Public Management reforms represent a case in point. Many or most of them were launched in order to increase efﬁciency, but interest in measuring the efﬁciency of the reforms – their actual effects on efﬁciency compared to their costs – seem to have been much smaller.

Actors or agents In summary, I have presented a picture of reformers that is considerably less heroic than the usual one. The reformers’ inﬂuence over the very existence of the reform is severely limited, as are their chances of determining its content. It is difﬁcult, moreover, to achieve the profound effects that reformers generally claim to be their goal. Reformers appear as agents rather than actors – as agents of a development that would have occurred with or without their intervention. Reformers are easily interchanged. Their position does not give them much power, primarily because organizations do not function as instruments for their management teams to the extent suggested by the institution of Organization, on which reforms are based. The aim of many reforms is to turn organizations into obedient instruments for their leaders. But reforms are difﬁcult to realize, simply because even in reform processes, this is not the way organizations work. None of these barriers to reform seem to curb the belief that reforms are key to organizational functioning. Nor do these barriers prevent ﬁerce conﬂict over who should be the reformers and how they should reform. This may not depend entirely on the fact that the idea of the ‘powerful reformer’ is taken seriously. People may consider reformers to be both signiﬁcant and valuable because they are aware of effects other than those discussed so far in this book. They may be considering the effects of reforms on people’s conceptions of reformers, of speciﬁc organizations, or of organizations in general – effects that are addressed in the next section.

134

Reform as Routine REFORM AS IMAGE BUILDING

The concept of reform refers to a relationship between thought and action, between ideas and practice; planning and implementing reforms shall reshape the practices of an organization in a certain intended way. The idea of reforming and the ideas that comprise the substance of a reform originate in an analysis of the problems and opportunities of the organization’s practice. The thinkers, the reformers, are to affect the doers, the reformees. In this book I have argued that the immediate connection between reform and practice may be a weak one, and that reforms belong more to the world of ideas than to the world of practice; sometimes they can be best understood only as ideas.

Reforming the reformers and the environment Rather than affecting practice, reforms may have effects in the world of ideas; they may affect people’s image of the organization. Through reform, people can acquire new ideas about the meaning and purpose of the organization and begin to think that they understand its operations better. When effects occur in the world of ideas rather than in the world of practice, actual reforms may violate another common assumption of reform: that the reformees and their activities are the sole target of a reform. Reforms may have signiﬁcant effects on other parties. Reforms may affect reformers – top managers, for instance. In large organizations, these are the people responsible for the organization and its operations. And yet they are as unable as anyone else to know everything that is happening in their organization – a situation that can lead to high uncertainty. The future organization described in a reform, on the other hand, is transparent and certain; it provides a model of the future organization that is relatively simple, easily understood, and, above all, under a higher degree of control than the present organization is. Details in this future organization can be changed with the stroke of a pen. Managers may ﬁnd it more attractive to be responsible for a simple and beautiful reform that they can easily grasp, and the content of which they seem to be able to design themselves rather than having to deal with

Reform and Power

135

the present, complex practice that functions inadequately, about which they have only partial knowledge and limited understanding, and over which they have limited control. Reforms make it easier to fulﬁl one of the primary tasks of top management: representing the organization to others. Rather than representing the actual, present organization managers can concentrate on representing the beautiful, future organization without having to deal with complaints from their subordinates about inaccurate descriptions. Reforms also affect attention. Both inside and outside the organization they direct attention towards the area being reformed and towards particular problems, values, and solutions raised by the reform. They draw attention to some especially problematic or unsatisfactory aspects of the organization’s practices. But reforms also provide hope for future improvements. They may even create the impression that improvements have already been made, as they are prone to generating talk of change among both reformers and reformees. The ultimate goal of most reforms is to improve the organization’s relationships with external parties. By changing forms, the organization’s operations should be more effective, efﬁcient, or service-oriented to the satisfaction of customers, the general public, or ﬁnanciers who become more willing to increase their support. But such satisfaction may be possible to achieve directly without detouring operations. The initiation of reforms demonstrates that the organization is dynamic and improving. Critical outsiders may become more tolerant of the current situation because they believe it will soon improve. Important outsiders with no ﬁrst-hand contact with daily operations – banks, stock market analysts, regulating states, or the media – may even confuse the reform with the present situation, believing that the organization already functions according to the reform that it has proclaimed. People who recognize the language of reform are likely to be more easily impressed than the general public is. In this case, it seems to be more difﬁcult for elites to manipulate the masses by mediating images that poorly reﬂect reality (Edelman 1971) than it is for them to convince one another of ideas in a highly abstract world where empirical aspects carry little weight. If a reform is meant to affect the opinions of external parties directly, it is essential to its success that its contents be made acceptable to these

136

Reform as Routine

parties. This is a further reason for adapting reform contents to general external norms and standards rather than to the speciﬁc situation of the organization to be reformed. A reform designed in this way is well ﬁtted for inﬂuencing outsiders, but less well adapted to internal implementation.

Change or stability In previous chapters, I have questioned the assumption that reform and change are closely related. On the contrary, reforms may contribute to stability. As described in Chapter 5, reforms are recurring phenomena; and in that sense, they express stability rather than change (reminding us of Sysophus – see Ku¨hl 2002). Reforms may even prevent change. Reforms increase people’s awareness that change is underway and activate resistance from those who oppose a speciﬁc change. And the reform project provides enemies of the reform with a forum in which they can oppose it. Controversy over reforms is likely to arise; reforms are attempts at achieving change by decision, and decisions draw attention to the fact that other options are possible and that people are responsible (Luhmann 2000). A reform decision makes it seem that the change is far from inevitable, that it results from the ideas of a limited number of decision makers, and that a decision not to reform or a decision to implement another reform could have been made. When controversies become strong enough, reforms fail. In such cases, reforms create stability. Actual change can only be achieved without reform, then. Such a situation is in accordance with the observation that many of the larger changes in society occur with little prior discussion, and that many of the most hotly debated areas result in little change (Edelman 1971). The potential, stabilizing effect of reform can make reform popular with the opponents of speciﬁc changes. Opponents can activate other people’s opposition to a threatening change by trying to represent it as a deliberate attempt by certain reformers to accomplish change. Another conceivable method of stopping a change could even be to represent it as one’s own idea for a reform. And, as noted in Chapter 5, proposing a new reform can be a way of preventing the full implementation of a reform that was introduced previously. The counter-strategy for those who do want change is to try to achieve it without reform.

Reform and Power

137

So reformers with controversial reforms may ﬁnd that their reforming ambitions actually counteract change. But this does not necessarily mean that the reformers will be hopelessly unpopular. Their efforts at reform show that they are trying to change things, which will be appreciated by those who want change. Moreover, the changes do not occur, which is comforting for those who do not want them. Failed reforms, then, provide a successful way of handling and responding to contradictory demands. Reform becomes a kind of organizational hypocrisy; reform talk about change can make stability in practice more acceptable. Perhaps the most important stabilizing effect of reforms is that they tend to reinforce prevailing ideas about organizational forms. Because reforms tend to be entrenched in generally accepted ideas, they may result in more people becoming – or pretending to become – convinced that these ideas are good and even that they are accurate descriptions of a given organization. Reforms also help to reinforce the more general image of how organizations work. Reforms reproduce images of organizations as systems that can be controlled from above, systems in which management or some other group of reformers possess great personal power and can bring about a direct change, and in which organizational forms represent important instruments of change. Yet the result of reforms poses a potential threat to this picture of the organization as controllable and possible to reform from above. If it becomes clear that a reform has failed, the idea of powerful management may be called into question. But as exempliﬁed in the next chapter, there are various ways to avoid this threat, ways of maintaining the hope that all aspects of the true organization can be realized in the future. Another possibility is that the leaders who have failed at reform resign, thus conﬁrming the idea that management is key. A third possibility is to formulate reforms in such a way that they cannot fail, perhaps by adapting their content to changes which would have taken place anyway, or to changes that appear to be attainable.

The power of perceptions If reforms affect the prevailing image of individual organizations and of organizations in general, then their implications for power are crucial. When

138

Reform as Routine

reforms project an image of powerful reformers and of change as something controlled by individuals, they are placing responsibility for the actions of organizations onto the reformers – whether that responsibility constitutes credit or blame. As reformers are managers or have the support of management, reforms also help to endow management with responsibility. Reforms represent one way for leaders to demonstrate their own importance, and therefore their responsibility, to the organization’s development. Their aspirations to responsibility may persuade them to choose explicit reform rather than trying to change things more discreetly, or they may decide to introduce reforms coinciding with changes that are already underway. The power and responsibility of managers is a fundamental notion that constructs explanations of what we see happening around us, and seems to provide us with instruments for change. By undertaking reforms, therefore, managers can reinforce their own responsibility, while inﬂuencing other people’s ideas about how the organization works. They can nurture an illusion of hierarchy and rationality that generates meaning and an incentive to act. But the unrealistic element in this illusion can lead to powerlessness in those who believe in it too strongly, whether they are the managers, the managed, or the outsiders. Those who want to change or inﬂuence the activities of an organization are misled about how to do it, believing that change is best effected by replacing the leaders or by making reforms. If those who want to change do not understand how to achieve it, this too promotes stability. When reforms inﬂuence images rather than practice, their paradoxical result is that the image of formal organizations as instruments and the image of reformers as heroes will be reinforced, just because they are not accurate. Reforms, then, simply obstruct any experiential learning. Reform activities will then continue to attract sizeable resources into administration, into ‘managing’ organizations by adopting the ‘right’ forms – even though the aim of many individual reforms is said to be just the opposite: to reduce administrative costs. In many organizations, huge administrative resources are sunk into more or less continuous reform activities, the cost–beneﬁt balance of which is seldom calculated. Higher administrative costs beneﬁt the administrators, but the beneﬁt they bring to others is more questionable. In summary, I have drawn a picture of reformers as people with considerably less power than the idea of reform implies. The limitations

Reform and Power

139

refer not only to what reformers can do, but even more to what they can say and want, and perhaps even to what they can think (cf. Douglas 1986). Organizational reform is determined to a large extent by external rules and general norms and ideas; the reformer’s prison walls are mental rather than physical. The most signiﬁcant of these rules and values is the very idea that organizations can easily be reformed and that reforms lead to strong positive effects on operations. These ideas are difﬁcult to abandon, not only for reformers but also for all those who want to live in the world of formal organizations.

8 The Hopeful Organization Now hope that is seen is not hope. For who hopes for what he sees? But if we hope for what we do not see, we wait for it with patience. Romans 8:24–5

Reforms are expressions of hope – of the hope that it will be possible to transform ideas and principles into practice. Via reforms, principles for how an organization ought to function shall govern its actual functioning. Hope is a necessary condition for reform. Without hope, reforms lose their meaning. Hope is a combination of prediction and desire (Capps 1995). Reforms are threatened by its alternatives: despair and apathy. Despair occurs if we do not believe that what we hope for will ever be attained, and apathy occurs when we have abandoned our wish for it to be attained. Reforms are also occasions, however, for people to run a high risk of losing hope – a risk created by the fact that reforms force the world of ideas to be confronted by the world of practice. On other occasions it is not difﬁcult to separate these two worlds, ignoring any inconsistency between them. Daily we talk one way and act in another (March 1986). On some occasions, organization members represent their organization in line with generally shared ideas of a true and appropriate organization, but these representations do not necessarily affect the way they act on other occasions. Or a division of labour upholds the separation between ideas and practice. Top management or the public relations department represents the organization as it ought to be to the outside world, while other organization members conduct operations without any one of them knowing much about the others or having much effect on the others. It is

The Hopeful Organization

141

easy to hope for lofty ideas and principles as long as we need not confront them with practice. In connection with reforms, however, the situation is different. Reforms force people to do something they rarely do: talk about ideas and practice at the same time in order to make them consistent. The goal is for consistency to be achieved by changing practice to be in accordance with ideas. People are willing to risk their practice (that is the point of reform), but they also risk their ideas, for it may be their ideas rather than their practice that change. It is precisely when we try to turn our ideas into practice that we are likely to discover that they are not easily translated into practice, that they are not appropriate for the practice we want to reform, or that they are completely unrealistic. This can happen early in the reform process, when people try to transform an attractive, general standard to more concrete guidelines for their own organization. Or with attempts at implementation and contact with reformees, reformers may discover that the attractive, simple, and self-evidently correct reform principles turn into a much more complex programme when it is adapted to actual, practical situations – a programme which no longer seems selfevident. Or a ﬁnal failure with a reform could lead to doubts about one’s initial beliefs – even to their rejection. Connecting our beliefs to practice in reforms, therefore, forces us to run the risk of losing hope that certain of our ideas can or will ever be realized. And if that happens, we are unlikely to continue reforming on the basis of these ideas. Our reforms then undermine chances of further reforms. But even in situations of reform, people may be able to maintain their hopes. Just as reforms sometimes leave practice unaffected by lack of implementation, they may leave ideas unaffected as well. In this chapter, I discuss how people can maintain their hope, even in connection with reforms. Such an ability increases the likelihood that they support further reforms. The frequency of reform is stimulated by a supply of hope.

THE HOPE FOR THE TRUE ORGANIZATION As St Paul argued in his letter to the Romans, it is easier to maintain hope when issues have been realized than when we are ignorant of the realism of

142

Reform as Routine

the issues. One type of reform that seems to require a great deal of effort for maintaining hope is the one intended to establish the ‘true’ or ‘proper’ organization. As described in previous chapters, such reforms are designed to create an organization characterized by a clear and strong identity, hierarchy, and rationality – not only in its representations, but also in practice. In attempts at reinforcing identity, reformers try to make the task of the organization or its business idea clearer, more coherent, and often simpler; they try to improve the hierarchy by installing better control and coordination systems, often by changing or expanding techniques for accounting, monitoring, and evaluation; and they try to improve rationality by giving the organization clearer goals and objectives or better budget and planning techniques. Such reforms are common, and the principle of the true organization has been part of many standards that have been or continue to be popular – management by objectives, for example, or process re-engineering or quality systems. Such reforms are particularly easy to sell. They are attractive. It is difﬁcult to reject the intention of making the organization more like a true organization. Yet the reforms are difﬁcult to implement. It is difﬁcult to make practice consistent with these strong formulations of identity, hierarchy, and rationality. Most of these reforms, in fact, seem to fail. If this were not the case, it would be difﬁcult to explain why there are so many such reforms. If they succeeded, the demand for them would be low, because most organizations would already be reformed into true organizations. The number of reforms would also be small if failure led to the loss of hope. It is only the combination of failure and retained hope, in fact, that can explain repetition of a reform with the same contents. The large number of reforms aimed at establishing the true organization is a sign not only of common failures, but also of the maintenance of hope in the face of failure. A situation in which reforms towards the true organization are repeated in the same organization signiﬁes that previous reforms have had little effect on either ideas or practice. If the reforms had, in fact, been implemented, there would be no need for further reforms; and if people had given up hope, there would be no wish to reform. So a complete explanation of reform requires an explanation of how people can maintain their hope, and how they can do so even in the face of discouraging experiences. How can we continue to believe in lofty ideas and principles that are not reﬂected in our practice? And why do reforms

The Hopeful Organization

143

not make those involved lose hope? How can they maintain their hope that the ideal of the true organization is realistic and useful and that it can be realized even when that ideal is confronted with organizational practice? Why do reforms not undermine future reform? Why don’t reformers give up? These are questions discussed in this chapter.

Maintaining hope In the next section, I present a selection of mechanisms for maintaining hope for the true organization. The mechanisms were revealed through extensive empirical studies of managers, other employees, management consultants, and politicians involved at different stages in reform attempts aimed at the realization of some aspects of the true organization. These people, all of whom were afﬁliated with ﬁrms or public agencies, played the roles of reformers, reformees, or observers of reform, and they all succeeded in preserving their hope even in reform situations.1

M EC H A N I S M S O F H O P E When our hope is threatened by our experience, we can defend it only by the avoidance of learning from experience or by learning from it in a special way. There were three fundamental ways in which the people studied in this research handled experience so that it did not threaten their hope that their lofty principles could be turned into practice. They were able to avoid confrontation between their principles and their experience in spite of the fact that they were dealing with reforms. Or they were able to select practice – to consider only practices that did not threaten their hope for their principles. Or when confronted with practical experience that threatened their hope, they were able to interpret the experience in a way that averted the threat. The next three subsections provide a few examples of each strategy. Some can be used in combination with others, and some represent alternatives to each other. 1 More mechanisms, as well as empirical illustrations and a thorough theoretical discussion, can be found in Brunsson (2006).

144

Reform as Routine

Avoiding practical experiences Confrontation with practice could be avoided by an interest in talk rather than practice and by a long-time perspective. Maintain interest in talk. As noted in Chapter 1, reforms are a special way of trying to accomplish change. Reformers try to change things by talking about them, by making a reform decision describing the new system to be installed. At the beginning, reforms consist only of talk, but the purpose is that talk will be transformed into practice. In the cases studied, however, it turned out that many reformers and reformees never became interested in practice. They were interested only in talk from the beginning to the end of the process. It is no simple matter to translate general ideas to more speciﬁc ones, into concrete ideas about what should be changed. Such fashionable standards as management by objectives, quality systems, or process reengineering entail general principles, and it is far from obvious how they should be applied in speciﬁc organizations. The reforms that have been studied typically started with a great deal of conceptual work by which people tried to create a reform scheme for their organization. Some reformers lost sight of practice in their fascination with the logic of this scheme. They spent all their time developing a logical reform model. They created documents in which they tried to write down objectives, strategies, and the like, in the right order. They stopped talking about practical effects, and switched to logic. And when they were ﬁnally satisﬁed that they had the logic right, they interpreted that as a sign that the reform was possible to implement and that it was valuable. Other people involved in reforms talked about effects, but effects on people’s thinking rather than practical effects. They argued that organization members had started to think according to the reform model, using the same concepts. In some cases, the reform was considered to have led to an ‘intellectual leap’ among reformees. This alleged effect was enough to lead to the conclusion that the reform was a great success. The reform could be evaluated without reference to practice, so practical experience could not undermine hope. Many managers found the reform model useful for representing their organization to the outside world. Top managers thought that the reform was a success because it provided a simple, comprehensible, and attractive

The Hopeful Organization

145

description of the organization. They used the model for representing their organization at international conferences. Many managers at lower levels of the organization reported that they did not use the reform model in their practice because it did not ﬁt there. They did say, however, that they found the reform model a handy tool for describing their department to other people, such as job applicants and newly hired employees. The actual situation was perceived as extremely complex, confused, and embarrassingly difﬁcult to describe. The reform model offered a clear, logical, and consistent image of the organization and a much better ﬁt for representation. All these people evaluated the reform on the basis of its effects on beliefs or the representations of their organization. And they reconﬁrmed that to think and talk in terms of the true organization is a valued way of thinking and talking. They had no reason to abandon hope that the ideas expressed in the reform were realistic, because they avoided confrontation between the principles and the practice of the reform and because their experience with the reform was a positive one. This mechanism of hope implies that people need not wait for the implementation of a reform in order to judge its success. They know long before implementation that it is a success. The classical problem of reform implementation merely evaporates. Think ahead. Reforms imply talk about the future, about future practice. Today’s practice is relevant only as a background and a motive for change. One way of avoiding contact with actual practice throughout the reform process was to maintain interest in the future and disinterest in the present. In that way, any practice that could be observed after the initiation of the reform was extraneous. The argument that a reform had had almost no practical consequences or that it had had unintended consequences could be refuted as irrelevant. It was the future consequences that counted. And the future can always be positive because it can be constructed without being tied to practice. Indeed, this future could be found in the formulation that was conveyed by the reformers at the beginning of the reform; it was the future of the true organization. The future looked bright when the reform began, and the future continued to look bright. Sometimes the organization members in this study managed to keep the bright future at a great distance. It was common for them to argue that

146

Reform as Routine

it would take many years before the reform could be expected to have an effect. I even found reformers who argued that the real consequences of their reform would not appear for twenty years. So lack of effects or unintended effects in the present did not indicate that the true organization would never be realized. There was no reason to lose hope. After some years, reforms are likely to be all but forgotten, as they are replaced by another reform (Brunsson, K. 1998). But even long memories need not threaten hope. After some forty years of trying to install models of more rational budget techniques oriented to objectives and results in the OECD countries, the OECD reported serious failures, but maintained hope by arguing that a ‘long-term approach and persistence are needed’ (OECD 2005). As St Paul has told us, patience is a close ally to hope (see also Capps 1995). One problem with the future is that over time it tends to become the present, and if there are still no positive effects of a reform when the predicted time has come, hope may vanish. But many reformers and reformees managed to avoid this aspect of the future. The future turned out to be a moving target. When the future became the present, there was always a new future to be considered. In one extreme case, I could ﬁnd no one who reported a single important positive effect of a reform after eight years of implementation. Yet, no one involved had lost hope. They saw no problem with the reform principle. A common response was, ‘We just have to wait a few more years’. In fact, this mechanism allows every confrontation with practice to be postponed, assisting hope in maintaining itself.

Select practice Sometimes people cannot completely separate themselves from knowledge about practice or about practical problems. But they can still be selective, choosing to consider only practices that fail to challenge the reform principle. This selection can be achieved in a variety of ways. Do not learn from your own practice. Reformers are always at some distance from the practice of the organization under reform. Yet there is one practice about which they are more knowledgeable: their own. Not surprisingly, that practice – like most other practice – turned out not to be

The Hopeful Organization

147

in line with high and lofty principles. All the reformers in this research failed to practice what they preached. So in order to maintain their hope, they needed to avoid a confrontation between their reform principles and their own practice. All of them managed to do so. One ﬁnancially successful consultant sold a particular reform programme, emphasizing that his clients’ organizations should be rationalized in line with one overall purpose: to satisfy the organizations’ customers and their needs. But when this consultant’s own clients wanted even minor changes in the standard implementation programme of his model, they were denied any ﬂexibility. He had his ready-made product, and his business concept was ‘take it or leave it’. The business concept of not adapting to client needs turned out to be a successful way for the consultant to generate business. But had the consultant learnt from his own practice, he would have lost his belief in the model he was selling. Other reformers had established detailed rules for introducing management by objectives – a standard in which the main idea is that organizations are best controlled from the top by management formulating general goals and objectives rather than detailed rules and commands; all micromanagement should be avoided. One such group of reformers published a 100-page booklet with detailed rules on how to proceed in order to avoid rules and become goal-oriented. In another handbook, consultants warned their clients against discussing the objectives of a management by objectives project with reformees before implementation, if at all. Although this strategy may well be a good one for initiating and implementing reform, it opposes the fundamental notion behind management by objectives. It is not particularly surprising that reformers who wanted organizations to change and implement reforms did not follow their own models, because such behaviour would have been unlikely to further their interests of selling and implementing reforms (Brunsson 2000). More surprising is their ability to isolate the experience of their practice from their product. This isolation prevented them from doubting their belief in the true organization. It prevented them from losing hope. Reforms are for others. Although, according to their standards, the consultants needed to reform their own practice, they only saw the need for their clients to reform. This attitude was also common among top managers. The managers who had decided upon reform were poor

148

Reform as Routine

followers of even the initial steps of the reform, in spite of the fact that top management had a crucial role to play, according to the reform models and to other reformers. Yet their own lack of adherence did not lead managers to doubts or reduced hope. Reformees, for their part, argued that reforms were more relevant for others than they were for themselves. Even when reformees reported that they did not and could not use the reform model in their own departments, they did not consider it to be a poor model. It was useful for others, not least for the other departments in their organization. Other departments had a great need to become rationally organized in accordance with the reform content. Again, by concentrating on the practice of others rather than their own practice, people could maintain their hope that the true organization is a realistic and useful project. If this is a typical mechanism – if we all tend to think that the true organization is good for others but not for ourselves – we can expect a greater supply of such reforms than demand for them. Indeed, there seem to be many eager reformers and many unwilling reformees. There have been ample reports in the literature of ‘resistance to change’ or ‘implementation problems’ (Winter 2003; Drazin et al. 2004). These phenomena may be a result of this particular mechanism of hope. Belief in change. Another way to learn about the practical effects of reforms is through the practice of other organizations that have tried to implement the same type of reforms. In my studies, I did encounter reformers who tried to learn from others. They ran the risk of discovering that the practices of other organizations had not changed in the intended way as the result of their reforms. They avoided this risk, however, by trying to learn from others who had not progressed far enough in their reform processes to notice any practical consequences. This selection of practice had to do with the reformers’ belief in change. The reformers wanted to introduce the latest reform concept in their own organizations. They saw no reason to introduce old ideas if new and allegedly better ones existed, so only experiences with new concepts were perceived to be relevant. They contacted other organizations that were working with the same new concepts, but because these concepts were new, these organizations had barely begun their reform processes, and had no practical consequences to report. So the knowledge-seeking reformers could not learn about practical difﬁculties, but only about the

The Hopeful Organization

149

ways in which other organizations talked about their reform concepts. And all the people they listened to seemed satisﬁed – as one might expect them to be at the beginning of a reform project (see Chapter 5). It was the interpretation that the new concept was really new that prevented the reformers from learning from others. Had they interpreted the ‘new’ concept as I do – as a slight variation of the same old idea of the true organization – they could have learned from the experience of many others; there is after all an enormous number of organizations that have tried to reform according to this idea. But these reformers believed in change. Because they saw their reform ideas as new, there was little room for learning. Belief in change, then, prevents learning and becomes a mechanism of hope. It seems that many sellers of reforms and reform concepts try to reinforce this mechanism. They argue that their reform concept is truly new and substantially different from any previous idea. In this way they help people to avoid learning from the past and to maintain their hope.

Interpreting practice In spite of all these attempts at avoiding and selecting information about practical problems, such information sometimes reached the reformers and reformees. The strategy for maintaining hope, then, was to interpret the information in a way that was not threatening to the principle – by looking on the bright side. There are several ways of doing that, three of which I mention here. Belief in change (again). It was not merely reform concepts that were seen to be changing; the perception of change could be extended to the world in general. And if the world was changing, experience was irrelevant. Even reports about failures in reforms did not serve as a real threat to hopes for the future, because these failures had occurred under circumstances that no longer existed or at least would not exist in the future. Reforms are, by deﬁnition, about the future! A multitude of changes would occur, and the idea of the true organization would be possible to realize this time, even if that had never been possible before. The world was moving in favour of these ideas. In one version of the story, the very act of starting to reform would change the circumstances in favour of

150

Reform as Routine

reform implementation. In this way, the ﬁrst reform steps would pave the way for later steps. Again, a belief in change served as a mechanism of hope, not only to prevent people from ﬁnding relevant information, but also to help them interpret existing information in a hopeful way. Some reformers tried to help with this mechanism by describing the world in ﬂux, characterized by turbulence. This mechanism of hope is an example of how a perception of change stabilizes rather than destabilizes the world. If people believe the world to be stable, they will be able to learn from experience and be more likely to conclude that their fundamental ideas are problematic and they will be more likely to lose hope. Such a development would represent a dramatic change. But if people think that the world is changing, they are able to maintain their ideas and their hope. An impression of change prevents change and leads to stability. Be ﬂexible. Many people in the organizations studied were able to adapt the objective of the reform, their aspiration level, and their expectations to what had actually been achieved by the reform. When launching the reform, they described it as involving a dramatic change and improvement; it would make the organization function in a completely new way. Yet little was achieved, even after a year. The only result that typically occurred within a year was the writing of some detailed documents. But by this time people described their initial purpose as being exactly what they had achieved. They now argued that they had had a modest purpose in the ﬁrst phase of the reform – that they had merely wanted people to learn the terminology and to use that terminology for writing certain texts. Because they had achieved that goal, the reform was successful so far. Put differently, even if the reform failed according to the initial objectives, it succeeded according to the new ones. In this way, people turned failure into success. And because the reform had been successful during the ﬁrst year, there was no reason to lose hope. It would continue to be successful. The fundamental mechanism was ﬂexibility – the ability to reduce ambitions over time. But people were ﬂexible in the other direction as well; they did not stick to these modest ambitions. Their ambitions were valid only for what was now deﬁned as the successful ﬁrst step. When they again looked to the future, they were able to revive and re-establish their original purpose, aspirations, and expectations. The purpose of continuing the next step of the reform work was again to achieve dramatic change.

The Hopeful Organization

151

They expected this dramatic change to happen in the future, just as they had expected at the initiation of the reform. Great hopes were revived. This level of ﬂexibility allows success to be combined with repetition and continuation. If the reformers had not been able to adapt their ﬁrst ambitions to their actual achievements, the reform would have proven a failure, and failure could have led them to lose hope and to stop any further attempts at reform. And if they had not been able to revive their old ambitions, there would have been no reason to continue reforming, because their ambitions would already have been reached. We are special. Some people in the study had to admit that they or their organizations had failed. They had not been able to implement their proposed reform. In this situation, there was a clear risk that they would conclude that the reform principle was impracticable or useless – that they would lose hope. One way they avoided that conclusion was to argue that they and their department or organization constituted special cases. Admittedly, the reform principles did not work. But that did not mean that the principles were fundamentally wrong. The principles were realistic and useful in other cases – in cases that were not special, in cases unlike their own case. Other cases were general rather than special, and people in general could, of course, beneﬁt from general principles because general principles ﬁt the general case. There is hope for them. Top managers were the people deciding to initiate the reform model, and they typically argued that it was important to install this model in their organization. But when, more often than not, they had problems applying the model themselves, those problems were not signs of something wrong with the model. It was their belief that top management constitutes a special operation, particularly in their company. Similarly, professional reformers argued that they did not operate by using their own reform models because trying to implement reforms is such a special activity. In short, people did not generalize from their own cases. The general rule was valid, even if it did not ﬁt their personal experience. This line of reasoning is supported not only by the warning of science against generalization (see Chapter 2), but also by the idea of individuality. As modern individuals, we seem to have one thing in common: seeing ourselves as special or even unique. With the help of this mechanism, personal experiences can be treated as irrelevant to general truth – as

152

Reform as Routine

deviations and exceptions to general truth. In this way, the truth can be maintained even if it does not ﬁt the experience of many people. The more special we consider ourselves to be, the more hopeful we can be. For organizations as a whole, uniqueness also constitutes a signiﬁcant value. Organizations must claim a special purpose, skill, or operation if they want to convince us that they have the right to exist as a separate entity. This notion of uniqueness renders the reform failures at one’s own organization less relevant, thereby supporting hope.

MAINTAINING HOPE By applying these and other mechanisms, people succeeded in maintaining their hope about the true organization when it was put to the test in reform situations. Maintaining hope made it possible to launch new reforms in the same direction. Or perhaps the mechanisms were not what made people hopeful. Rather, they may ﬁrst have chosen to hope and then used the mechanisms to defend this position. It is difﬁcult to say. But people have reason to sustain their hope. Losing hope would be costly, not only because hopefulness is a more positive state of mind than is hopelessness. Losing hope would also be in conﬂict with the general perspective of a hopeful Western culture. Hope is fundamental in our culture because it supports other fundamental ideas. The existence and robustness of hope contributes strongly to the situation I described in Chapter 2 – our ability to exist in two worlds simultaneously: a world of ideas and a world of practice. It also allows us to adhere to the Platonic heritage of perceiving the world of practice as a fallen world, deviating from what we know is true and right. Hope makes the ‘true’ world seem realistic, albeit only in the eschatological sense that it can be achieved in the future. The modern, secular version conceives of this future as not too distant, as deﬁnitely on this earth, and as achievable by reforms. Declaring a lack of hope in this cultural situation is likely to lead to awkward questions and the need to defend a deviant position on other fundamental issues. One may be required to refute the realism of lofty ideals, be unable to emphasize the future, and question the idea of progress.

The Hopeful Organization

153

Organizations are environments in which it may be critical, even required, that one be hopeful. Organizations are projects, the very existence of which expresses a certain level of hopefulness. Organizations are attempts at achieving things – not ﬁnal results (Ahrne and Brunsson 2008, Ch 3). Most organizations have extensive experiences with failure – with things not going according to plan. But is it not meaningless to organize and to mobilize organizational action if there is no hope of future success? It is difﬁcult to organize on the basis of despair or apathy. A certain degree of hopefulness seems to be a component of our role as members of organizations. Although there are organizations for almost anything and anybody in the modern world, it is difﬁcult to believe that there are many associations of apathetic people.

Continued reforming In summary, and seemingly paradoxically, hope stabilizes our beliefs, maintains our ability to separate the world of ideas from the world of practice, and therefore stimulates reform. The discussion in this chapter does not imply that people never lose hope, not even in organizations. But there are enough mechanisms of hope to make it possible and even common that hope is retained, even under experiences of failed reforms and even for such ambitious hopes as the hope for the true organization. The failed reform accompanied by retained hope is fertile soil for new reforms. And as previous chapters have demonstrated, there are other dynamics of reforms in general that tend to produce new reforms. Organizations will continue reforming. The study of organizational reform should continue as well.

This page intentionally left blank

References Abbott, A. (1988) The System of Professions. An Essay on Division of Expert Labor. Chicago: The University of Chicago Press. Abernethy, Margaret A. and Stoelwinder, Johannes U. (1995). The Role of Professional Control in the Management of Complex Organizations. Accounting, Organizations and Society, 20/1: 1–17. Abrahamson, Eric (1996). Management Fashion. Academy of Management Review, 21: 254–85. Ahrne, Go¨ran (1994). Social Organizations. Interaction Inside, Outside and Between Organizations. London: Sage Publications. —— (1998). Stater och andra organisationer. In Ahrne, G. (ed.): Stater som organisationer. Stockholm: Nerenius and Sante´rus fo¨rlag. —— Brunsson, Nils (2008). Meta-Organizations. Cheltenham: Edward Elgar. Alvarez, J.-L. (1991). The International Diffusion and Institutionalization of the New Entrepreneurship Movement. A Study in the Sociology of Knowledge. Harvard University (Diss.), Boston. Alvesson, M. and Berg, P. O. (1992), Corporate Culture and Organizational Symbolism. Berlin/New York: Walter de Gruyter. Andersson, A˚., Fu¨rth, T., and Holmberg, I. (1993). 70-talister, om va¨rderingar fo¨rr, nu och i framtiden. Stockholm: Bokfo¨rlaget Natur och Kultur. Andersson, U., Johanson, J., and Vahlne, J. (1997). Organic Acquisition in the Internationalization Process of the Business Firm. Management International Review, 37/2: 67–84. Argyris, D. C. and Scho¨n, D. (1978). Organizational Learning: A Theory of Action Perspective. Reading, Mass: Addison-Wesley. Aristotle (1) (384–322 bc) (1985). Nicomachean Ethics. Indianapolis, IN: Hackett. —— (2) (384–322 bc) (1984). The Politics. Chicago: University of Chicago Press. Aucoin, Peter (1990). Administrative Reform in Public Management: Paradigms, Principles, Paradoxes and Pendulums. Governance 3/2: 115–37. Axelsson, B. and Easton, G. (1992). Industrial Networks – A New View of Reality. London: Routledge. A˚kerstrøm Andersen, Niels (1995). Selvskabt forvaltning: forvaltningspolitikkens og centralforvaltningens udvikling i Danmark fra 1900–1994. Copenhagen: Nyt Fra Samfundsvidenskaberne.

156

References

Baier, V. E., March, J. P., and Sætren, H. (1986). Implementation and Ambiguity. Scandinavian Journal of Management Studies, 2/3–4: 197–212. Barnard, Chester (1938). The Functions of the Executive. Cambridge, MA: Harvard University Press. Berg, Per Olof and Jonsson, Christer (1991). Strategisk ledning pa˚ politiska marknader. Lund: Studentlitteratur. Berger, P., and Luckmann, T. (1966). The Social Construction of Reality: A Treatise in the Sociology of Knowledge. New York: Doubleday. Bergeva¨rn, L. (1989). Processer och krafter kring fo¨retags och kommuners omformning av sin redovisning. Unpublished paper. Berman, M. (1982). All that is Solid Melts into Air. The Experience of Modernity. George Borchardt. Berrevin, Renaud and Musselin, Christine (1996). Les politiques de contractualisation entre centralisation et de´centralisation: Les cas de l’e´quipement et de l’enseignement supe´rieur. Sociologie du Travail, 4/96: 575–96. Boston, Jonathan (ed.) (1995). The State under Contract. Wellington: Bridget Williams Books. —— Martin, John, Pallot, June, and Walsh, Pat (1996). Public Management: The New Zealand Model. Oxford: Oxford University Press. Bourdieu, Pierre (1984). Distinction: A Social Critique of the Judgement of Taste. Cambridge, MA: Harvard University Press. Brunsson, Karin (1995). Dubbla budskap. Hur riksdagen och regeringen presenterar sitt budgetarbete. Stockholm: Stockholms universitet, Fo¨retagsekonomiska institutionen. —— (1998). Non-Learning Organization. Scandinavian Journal of Management, 14/4: 421–32. —— (2002) Organisationer. Lund: Studentlitteratur. Brunsson, Nils (2000). The Irrational Organization. 2nd edn. Bergen: Fagbokforlaget. (1st edn. 1995 John Wiley & Sons). —— (2002). The Organization of Hypocrisy. Talk, Decision and Actions in Organizations. 2nd edn. Oslo: Copenhagen Business School Press. (1st edn. 1989 by John Wiley & Sons). —— (2006). Mechanisms of Hope. Maintaining the Dream of the Rational Organization. Malmo¨: Copenhagen Business School Press. —— (2007). The Consequences of Decision-Making. Oxford: Oxford University Press. —— and Ha¨gg, Ingemund (eds) (1992). Marknadens makt. Stockholm: SNS fo¨rlag. —— and Olsen, Johan P. (1997) The Reforming Organization 2nd ed. Bergen: Fagbokforlaget (1st edn. 1993 by Routledge).

References

157

—— —— (1998). Organization Theory: Thirty Years of Dismantling, and then . . . ?. In Brunsson, N. and Olsen, J. P.: Organizing Organizations. Bergen: Fagbokforlaget. —— and Jacobsson, Bengt (2000). A World of Standards. Oxford: Oxford University Press. —— Forssell, Anders, and Winberg, Hans (1989). Reform som tradition. Administrativa reformer i Statens Ja¨rnva¨gar. Stockholm, EFI. Burns, T. and Stalker, G. M. (1961). The Management of Innovation. London: Tavistock. Callon, M. (1998). Introduction: The Embeddedness of Economic Markets in Economics. In Callon, M. (ed.): The Laws of the Market. Oxford: Blackwell. Capps, Donald (1995). Agents of Hope: A Pastoral Psychology. Minneapolis: Fortress Press. Christensen, Tom (1991). Virksomhetsplanlegging. Oslo: Tano. Coase, R. H. (1937). The Nature of Firm. Economica, 5: 386–405. Cohen, M., March, J. G., and Olsen, J. P. (1972). A Garbage Can Model of Rational Choice. Administrative Science Quarterly, 1: 1–25. Colado, Eduardo (1996). Excellence at Large: Power, Knowledge and Organizational Forms in Mexican Universities. In Clegg, Steward R. and Palmer, G. (eds): The Politics of Management Knowledge. London: Sage. Crozier, Michel (1964).The Bureaucratic Phenomenon. Chicago, IL: University of Chicago Press. Cyert, Richard M. and March, James G. (1963). A Behavioral Theory of the Firm. Englewood Cliffs, NJ: Prentice Hall. Czarniawska, Barbara (1985). The Ugly Sister: On Relationships between the Private and the Public Sector in Sweden. Scandinavian Journal of Management Studies, 2/2: 83–103. Czarniawska-Joerges, Barbara (1988). To Coin a Phrase: On Organizational Talk, Organizational Control and Management Consulting. Stockholm: EFI. —— (1994). Narratives of Individual and Organizational Identities. In Deetz, S. A. (ed.): Communication Yearbook, 17: 193–221 Thousand Oaks, CA: Sage. —— Sevo´n, G. (eds) (1996). Translating Organizational Change. Berlin: Walter de Gruyter. Dahl, R. A. and Lindblom, C. E. (1953). Politics, Economics and Welfare. New York: Harper & Brothers. Danielsson, A. (1983). Fo¨retagsekonomi – en o¨versikt. Lund: Studentlitteratur. DiMaggio, P. J. and Powell, W. W. (1983). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. American Sociological Review, 48: 147–60.

158

References

Douglas, M. (1986). How Institutions Think. Syracuse, NY: Syracuse University Press. Drazin, Robert, Glynn, Mary Ann, and Kazanjan Robert, K. (2004). Dynamics of Structural Change. In Poole, Marshall Scott, and Van de Ven, Andrew H. (eds): Handbook of Organizational Change and Innovation. New York: Oxford University Press. Drori, G. et al. (2003). Science in the Modern World Polity. Institutionalization and Globalization. Stanford: Stanford University Press. —— Meyer, J., and Hwang, H. (2006). Organization and Globalization. World Society and Organizational Change. Oxford: Oxford University Press. Easton, Brian (1997). The Commercialisation of New Zealand. Auckland: Auckland University Press. Edelman, M. (1971). Politics as Symbolic Action. New York: Academic Press. Edwards, R. B. (1969). Freedom, Responsibility and Obligation. Haag: Martinus Nijhoff. Etzioni, A. (1964). Modern Organizations. Englewood Cliffs, NJ: Prentice Hall. Fayol, Henri (1916). Administration industrielle et ge´ne´rale. In Bulletin de la Societe´ de l’Industrie Mine´rale, Paris. Fernler, Karin (1996). Ma˚ngfald eller likriktning. Stockholm: Nerenius och Sante´rus fo¨rlag. Forssell, Anders (1990). Being Modern and Business-minded. In Co-operative Development and Change. Stockholm: The Society for Co-operative Studies. —— (1992). Moderna tider i Sparbanken. Stockholm: Nerenius och Sante´rus. —— Jansson, David (1996). The Logic of Organizational Transformation: On the Conversion of Non-business Organizations. In Czarniawska-Joerges, B., and Sevo´n, G (eds): Translating Organizational Change. Berlin: Walter de Gruyter. Freddi, Giorgio and Bjo¨rkman, James W. (1989). Controlling Medical Professionals. The Comparative Politics of Health Governance. London: Sage. Friedberg, Erhard. (1993). Le pouvoir et la re`gle. Dynamiques de l’action organise´e. Paris: Seuil. Fry, Geoffrey K. (1984). The Development of the Thatcher Government’s ‘‘Grand Strategy’’ for the Civil Service. A Public Policy Perspective. Public Administration 62, (Autumn): 322–35. Furusten, Staffan (1995). The Managerial Discourse – a Study of the Creation and Diffusion of Popular Management Knowledge. Uppsala: Fo¨retagsekonomiska institutionen. —— (2000). The Knowledge Base of Standards. In Brunsson, Nils and Jacobsson, Bengt: A World of Standards. Oxford: Oxford University Press.

References

159

Giddens, Anthony (1990). The Consequences of Modernity. Stanford: Stanford University Press. Grabher, G. (ed.) (1993). The Embedded Firm. On the Socioeconomics of Industrial Networks. London: Routledge. Granovetter, M. (1985). Economic Action and Social Structure: The Problems of Embeddedness. American Journal of Sociology, 91/3: 481–510. Ham, Christopher (1997). Health Care Reform. Learning from International Experience. Buckingham: Open University Press. Hanf, K. and Scharpf, F. W. (eds) (1978). Interorganizational Policy Making. London: Sage. Harrison, Stephen and Pollitt, Christopher (1994). Controlling Health Professionals. Buckingham: Open University Press. Hayek, F. (1988). The Fatal Conceit. Chicago: University of Chicago Press. Hellgren, B. and Melin, L. (1992). Business Systems, Industrial Wisdom and Corporate Strategies: the Case of the Pulp and Paper Industry. In Whitley, R. (ed.): European Business Systems: Firms and Markets in their National Contexts. London: Sage. Henning, Roger (2000). Selling Standards. In Brunsson, Nils and Jacobsson, Bengt: A World of Standards. Oxford: Oxford University Press. Hernes, G. (1978). Forhandlingsøkonomi og blandingsadministrasjon. Bergen: Universitetsforlaget. Holmblad Brunsson, Karin (2007). The Notion of General Management. Malmo¨: Copenhagen Business School Press. Hood, Christopher (1991). A Public Management for all Seasons? Public Administration, 69: 3–19. —— (1995). The New Public Management in the 1980’s. Variations on the Theme. Accounting, Organizations and Society, 20/2–3: 93–110. —— (1996). United Kingdom: From Second Chance to Near-miss Learning. In Olsen, J. P. and Peters, B. G. (eds): Lessons from Experience: Experiential Learning in Administrative Reforms in Eight Democracies. Oslo: Scandinavian University Press. Hopwood, Anthony G. and Miller, Peter B. (eds) (1994). Accounting as Social and Institutional Practice. Cambridge: Cambridge University Press. Humphrey, Christopher and Olson, Olov (1995). Caught in the Act: Public Services Disappearing in the World of ‘Accountable Management’? In Ashton, D., Hopper, T., and Scrapens, R. (eds): Issues in Management Accounting. Exeter: Prentice Hall. Huzcynski, G. (1993). Management Gurus – What Makes Them and How to Become One. London: Routledge.

160

References

Ha˚kansson, H. (1989). Corporate Technological Behaviour. Co-operation and Network Approach. London Routledge. —— Johansson, J. (1993). The Network as a Governance Structure. In Grabher, G. (ed.): The Embedded Firm. On the Socioeconomics of Industrial Networks. London: Routledge. Inglehart, R. (1977). The Silent Revolution, Changing Values and Political Styles Among Western Publics. Princeton: Princeton University Press. Jacobsson, Bengt and Sahlin-Andersson, Kerstin (1995). Skolan och det nya verket. Stockholm: Nerenius och Sante´rus fo¨rlag. Jepperson, R. (1991). Institutions, Institutional Effects, and Institutionalism. In Powell W. and DiMaggio, P. J. (eds): The New Institutionalism in Organizational Analysis. Chicago: University of Chicago Press. ¨ r kommunerna slavar under modetrenJohansson, P. and Johnsson, M. (1994). A derna?. Go¨teborg: Fo¨rvaltningsho¨gskolan, Go¨teborgs universitet. —— —— (1995). Kommunernas so¨kande i reformernas labyrint. Go¨teborg: Fo¨rvaltningsho¨gskolan, Go¨teborgs universitet. Jo¨nsson, S. A. and Lundin, R. A. (1977). Myths and Wishful Thinking as Management Tools. In Nystrom, P. C. and Starbuck, W. H. (eds): Prescriptive Models of Organizations. Amsterdam: North-Holland. Kaplan, A. (1964). The Conduct of Inquiry. San Francisco: Chandler. Krasner, S. D. (1999). Sovereignty: Organized Hypocrisy. Princeton: Princeton University Press. Ku¨hl, Stefan (2002). Sisyphos in Management. Die vergebliche Suche nach der optimalen Organisationsstruktur. Weinheim: Wiley. Kuhn, T. S. (1962). The Structure of Scientiﬁc Revolutions. Chicago: University of Chicago Press. Lamoreaux, Naomi R. (2004). Partnerships, Corporations, and the Limits on Contractual Freedom in U.S. History: An Essay in Economics, Law, and Culture. In Lipartito, Kenneth, and Sicilia, David B. (eds): Constructing Corporate America. History, Politics, Culture. Oxford: Oxford University Press. Lapsley, Irvine (1997). The New Public Management Diaspora: The Health Care Experiment. International Association of Management Journal, 9/2: 1–14. Latour, B. (1986). The Powers of Association. In Law, J. (ed.): Power, Action and Belief: A New Sociology of Knowledge? London: Routledge & Kegan. —— (1987). Science in Action. Milton Keynes: Open University Press. Laughlin, Richard (1996). Principals and Higher Principals: Accounting for Accountability in the Caring Professions. In Munro, R. and Mouritsen, J.: Accountability: Power, Ethos and the Technologies of Managing. London: International Thomson Business Press.

References

161

Lien, S. and Fremstad, J. (1989). Organisering av organisasjonsendring. In Egeberg, M. (ed.): Institusjonspolitik og forvaltningsutvikling. Oslo: Tano. Lindblom, Charles E. (1977). Politics and Markets. The World’s Political–Economical Systems. New York: Basic Books. Lindstro¨m, Stefan (1997). Det sva˚ra samspelet. Stockholm: SOU 1997:15. Luckmann, T. (1967). The Invisible Religion. New York: MacMillan. Luhmann, Niklas (2000). Organisation und Entscheidung. Opladen: Westdeutscher Verlag. Lu¨der, Klaus (1998). Towards a New Financial Management for Germany’s Public Sector: Local Governments Lead the Way. In Olson, O., Guthrie, J., and Humphrey, C. (eds): Global Warning! Debating International Developments in New Public Financial Management. Oslo: Cappelen Akademisk forlag. Malinowski, B. (1927). The Life of Culture. In Smith, G. E., Malinowski, B., Spinden, H., and Goldenweiser, A. (eds): Culture. The Diffusion Controversy. New York: W. W. Norton. Mandeville, B. (1724) (1966). The Fable of the Bees or Private Vices, Public Beneﬁts. Oxford: Oxford University Press. March, James G. (1981a). Decisions in Organizations and Theories of Choice. In Van de Ven, A. H. and Joyce, W. F. (eds): Perspectives on Organizational Design and Behavior. New York: Wiley. —— (1981b). Footnotes to Organizational Change. Administrative Science Quarterly, 4/26: 563–77. —— (1986). How We Talk and How We Act: Administrative Theory and Administrative Life. In Sergiovanni, Thomas J. and Corbally, John Edward (eds): Leadership and Organizational Culture: New Perspectives on Administrative Theory and Practice. Urbana: University of Illinois Press. —— Olsen, J. P. (1983). Organizing Political Life: What Administrative Reorganization tells us about Government. American Political Science Review, 77/2: 281–96. —— —— (1989). Rediscovering Institutions: The Organizational Basis of Politics. New York: Free Press. —— Simon, H. A. (1958). Organizations. New York: Wiley & Sons. Martin, Joanne (1992). Cultures in Organizations. Three Perspectives. Oxford: Oxford University Press. Martin, John (1995). Contracting and Accountability. In Boston J. (ed.): The State under Contract. Wellington: Bridget William Books. McSweeney, Brendan (1994). Management by Accounting. In Hopwood, A. G. and Miller P. (eds): Accounting as Social and Institutional Practice. Cambridge University Press.

162

References

Meyer, J. W. (1994). Rationalized Environments. In Scott, W. R. and Meyer, J. W.: Institutional Environments and Organizations. London: Sage. —— Jepperson, Ronald L. (2000). The ‘‘Actors’’ of Modern Society: The Cultural Construction of Social Agency. Social Theory, 18: 100–20. —— Rowan, B. (1977). Institutionalized Organizations: Formal Structure as Myth and Ceremony. American Journal of Sociology, 83: 364–85. Meyer, J. W. and Scott, R. W. (1994). Institutional Environments and Organizations. California: Sage. —— Boli, John, and Thomas, George M. (1987). Ontology and Rationalization in the Western Account. In Thomas, G. M., Meyer, J. W., Ramirez, F. O., and Boli, J. (eds): Constituting, State, Society and the Individual. Newbury Park, CA: Sage. —— Drori Gili, S., and Hwang, Hokyu. (2006). World Society and the Proliferation of Formal Organization. In Drori, G. S., Meyer, J. W., and Hwang, H. (eds): Globalization and Organization. World Society and Organizational Change. Oxford: Oxford University Press. Miller, Peter (1992). Accounting and Objectivity: The Invention of Calculating Selves and Calculable Spaces. Annals of Scholarship, 9/1–2: 61–86. —— (1994). Accounting as Social and Institutional Practice: An introduction. In Hopwood, A. G. and Miller, P. (eds): Accounting as Social and Institutional Practice. Cambridge: Cambridge University Press. ——O’Leary, Ted (1987). Accounting and the Construction of the Governable Person. Accounting, Organizations and Society, 12/3: 235–65. —— —— (1994). Governing the Calculable Person. In Hopwood, A. G. and Miller, P. (eds): Accounting as Social and Institutional Practice. Cambridge: Cambridge University Press. Mintzberg, Henry (1979). The Structuring of Organizations. Englewood Cliffs, NJ: Prentice Hall. —— (1994). The Rise and Fall of Strategic Planning. New York: Free Press. Mouritsen, Jan (1997). Taellelighedens regime. Copenhagen: Jurist och økonomforbundets forlag. Musselin, Christine (1997). State/University Relations and How to Change Them: The Case of France and Germany. European Journal of Education, 32/2: 145–64. Nahavandi, Afsaneh and Malakzadeh, Ali R. (1988). Acculturation in Mergers and Acquisitions. Academy of Management Review, 13/1: 79–90. Nilsson, Karin and A˚slo¨v, A. (1990). Vad a¨r privatisering? Stockholm: EFI, Working Paper. OECD (1995). Governance in Transition. Paris: OECD.

References

163

Olsen, Johan P. and Peters, B. Guy (1996). Lessons from Experience. Oslo: Scandinavian University Press. Olson, Olov (1997). Reform as a Learning Process. In Brunsson, N. and Olsen, J. P. (eds): The Reforming Organization, 2nd edn. Bergen: Fagbokforlaget. —— Guthrie, James, and Humphrey, Christopher (eds) (1998). Global Warning: Debating International Developments in New Public Financial Management. Oslo: Cappelen Akademisk forlag. —— and Sahlin-Andersson, Kerstin (1998). Accounting Transformation in an Advanced Welfare State. The Case of Sweden. In Olson, O., Guthrie, J., and Humphrey, C. (eds): Global Warning: Debating International Developments in New Public Financial Management. Oslo: Cappelen Akademisk forlag. Olsson, Lena (1995). Styra med kvalitet. FOU:rapport 1995:16, Stockholms socialtja¨nst. Ouchi, William G. (1977). The Relationship between Organization Structure and Organizational Control. Administrative Science Quarterly, 32/1: 264–89. Pfeffer, Jeffrey (1981). Power in Organizations. Marshﬁeld: Pitman. —— Salancik, Gerald R. (1978). The External Control of Orgnizations. New York: Harper and Row. Polanyi, K. (1968). Primitive, Archaic and Modern Economies. In Dalton, G. (ed.): Essays by Karl Polanyi. Boston: Beacon Press. Pollitt, Christoffer (1990) Managerialism and the Public Services. Oxford: Blackwell. —— (2009). Bureaucracies Remember, Post-bureaucratic Organizations Forget? In Public Administration, vol. 87, no. 2. Power, Michael (1994). The Audit Society. In Hopwood, A. G. and Miller, P. (eds): Accounting as Social and Institutional Practice. Cambridge: Cambridge University Press. —— (1997). The Audit Society: Rituals of Veriﬁcation. Oxford: Oxford University Press. Powell, W. W. and DiMaggio, P. J. (eds) (1991). The New Institutionalism in Organizational Analysis. Chicago: The University of Chicago Press. Pressman, Jeffrey L. and Wildavsky, A. (1973). Implementation. Berkeley: University of California Press. Preston, Alistair M. (1992). The Birth of Clinical Accounting: A Study of the Emergence and Transformations of Discourses on Costs and Practices of Accounting in U.S. Hospitals. Accounting, Organizations and Society, 17/1: 63–100. —— Cooper, David J., and Coombs, Rod W. (1992). Fabricating Budgets: A Study of the Production of Management Budgeting in the National Health Service. Accounting, Organizations and Society, 17/6: 561–93.

164

References

Reichard, C. (1995). ‘New Steering Models’ as the German Variant of New Public Management Concepts. Paper presented at the EGPA Conference, Rotterdam. Richardson, G. B. (1972). The Organization of Industry. The Economic Journal, 327/82: 883–96. Roberts, John (1991). The Possibilities of Accountability. Accounting, Organizations and Society, 16/4: 355–68. —— Scapens, Robert W. (1985). Accounting Systems and Systems of Accountability: Understanding Accounting Practices in their Organisational Contexts. Accounting, Organizations and Society, 10/4: 443–56. Rombach, Bjo¨rn (1991). Det ga˚r inte att styra med ma˚l. Lund: Studentlitteratur. —— (1997a). Den marknadslika kommunen. Stockholm: Nerenius och Sante´rus fo¨rlag. —— (1997b). Success at the Expense of Control. In Brunsson, N., and Olsen, J. P. (eds): The Reforming Organization. Bergen: Fagbokforlaget. —— Sahlin-Andersson, Kerstin (1995). Fra˚n sanningsso¨kande till styrmedel. Stockholm: Nerenius och Sante´rus fo¨rlag. Rothstein, Bo (1994). Vad bo¨r staten go¨ra? Om va¨lfa¨rdstatens moraliska och politiska logik. Stockholm: SNS fo¨rlag. Røvik, Kjell-Arne (1996). Deinstitutionalization and the Logic of Fashion. In Czarniawska-Joerges, B. and Sevo´n, G. (eds): Translating Organizational Change. Berlin: Walter de Gruyter. —— (1998). Moderne organisasjoner, trender i organisasjonstenkningen ved tusena˚rsskiftet. Bergen: Fagbokforlaget. —— (2007). Trender og Translasjoner. – Ideer som former det 21. a˚rhundrets organisasjon. Oslo:Universitetsforlaget. Sahlin-Andersson, Kerstin (1994). Varfo¨r la˚ter sig organisationer omvandlas? In Jacobsson, B. (ed.): Organisationsexperiment i kommuner och landsting. Stockholm: Nerenius och Sante´rus fo¨rlag. —— (1996). Imitating by Editing Success. In Czarniawska-Joerges, B. and Sevo´n, G. (eds): Translating Organizational Change. Berlin: Walter de Gruyter. Sales, Arnold L. and Mirvis, Philip H. (1984). When Cultures Collide: Issues in Acquisition. In Kimberley, J. R. and Quinn, R. E. (eds): Managing Organizational Transitions. Homewood, IL: R. D. Irwin. Samuelsson, P. (1964). Economics. An Introductory Analysis. New York: Prentice Hall. Schick, Allen (1996). The Spirit of Reform: Managing the New Zealand State Sector in a Time of Change. A Report Prepared for the State Services Commission and The Treasury, New Zealand.

References

165

Scott, W. Richard (1991). Unpacking Institutional Arguments. In Powell W. W. and DiMaggio, P. J. (eds): The New Institutionalism in Organizational Analysis. Chicago: The University of Chicago Press. Scott, R. W. (1998). Organizations. Rational, Natural and Open Systems. 4th edn. Englewood Cliffs, NJ: Prentice Hall. Segal, J. M. (1990). Alternative Conceptions of the Economic Realm. Institute for Philosophy and Public Policy. University of Maryland: Working Paper. Selznick, Philip (1949). TVA and the Grass Roots. CA: University of California Press. —— (1957). Leadership in Administration. New York: Harper and Row. Simon, Herbert (1945). Administrative Behavior. New York: Free Press. —— (1991). Organizations and Markets. The Journal of Economic Perspectives, 5/2: 25–44. Sjo¨strand, Sven-Erik (1985). Samha¨llsorganisation. Lund: Doxa. Smith, A. (1759) (1976). The Theory of Moral Sentiments. Oxford: Clarendon Press. —— (1776) (1976). An Inquiry into the Wealth of Nations. Oxford: Clarendon Press. Smith, G.E., Malinowski, B., Spinden, and Goldenweiser, A. (1927). Culture. The Diffusion Controversy. New York: W. W. Norton. Socialstyrelsen (1992). Vad menar man med kvalitet och hur bedo¨mer man kvalitet i barnomsorgen? Report. Sorge, Arndt (1996). Organizational Behaviour. In Warner, M. (ed.): International Encyclopedia of Business and Management, 4: 3793–3820. London: International Thompson Publishing. SOU (1997). I medborgarnas tja¨nst. Beta¨nkande fra˚n fo¨rvaltningspolitiska kommissionen. Stockholm. Spender, J.-C. (1989). Industry Recipes, an Enquiry into the Nature and Sources of Managerial Judgement. Oxford: Basil Blackwell Ltd. Starbuck, W. H. (1983). Organizations as Action Generators. American Sociological Review, 92–102. —— Nystrom, P. (1981). Designing and Understanding Organizations. In Starbuck, W. H. and Nystrom, P. (eds): Handbook of Organizational Design 1. New York: Oxford University Press. Stewart, John and Walsch, Kieron (1992). Change in the Management of Public Services. Public Administration, 70: 499–518. Strang, David and Meyer, John W. (1993) Institutional Conditions for Diffusion. Theory and Society, 22: 487–511. Sundstro¨m, Go¨ran (2003). Stat pa˚ villova¨gar. Stockholm: Stockholms Universitet.

166

References

Svedberg-Nilsson, Karin (1999). Effektiva fo¨retag. Stockholm: EFI. Swedberg, R. (1994). Markets as Social Structures. In Smelser, N. J. and Swedberg, R. (eds): The Handbook of Economic Sociology. Princeton, NJ: Princeton University Press. Tawney, R. H. (1926). Religion and the Rise of Capitalism. London: Joka Murray. Thomas, G. M., Meyer, J. W., Ramirez, F. O., and Boli, J. (1987). Institutional Structure: Constituting State, Society, and the Individual. Beverly Hills, California: Sage. Thompson, James D. (1967). Organizations in Action. New York: McGraw-Hill. Vad, Torben, B. P. (1998). Europeanisation of Standardisation. Institute of Political Science, University of Copenhagen. Veyne, P. (1983). Les Grecs ont-ils cru a` leurs myths? Paris: Seuil. Wallander, J. (1994). Budgeten – ett ono¨digt ont. Stockholm: SNS Fo¨rlag. Weber, Max (1924) (1947). The Theory of Social and Economic Organization. New York: The Free Press. —— (1925) (1968). Economy and Society. Berkeley, CA: University of California Press. Weick, K. E. (1969). The Social Psychology of Organizing. Reading, MA: AddisonWesley. —— (1976). Educational Organization as Loosely Coupled Systems. Administrative Science Quarterly, 21: 1–19. Westney, D. E. (1987). Imitation and Innovation: the Transfer of Western Organizational Patterns to Meiji Japan. Cambridge: Harvard University Press. Williamson, Oliver. E. (1975). Markets and Hierarchies: Analysis and Antitrust Implications. New York: The Free Press. —— (1981). The Modern Corporation: Origins, Evolution, Attribution. Journal of Economic Literature, 1537–68. Winter, Søren C. (2003). Implementation Perspectives: Status and Reconsideration. In Peters, B. Guy and Pierre J. (eds): Handbook of Public Administration. Thousand Oaks. California: Sage. Woodward, J. (1965). Industrial Organization: Theory and Practice. Oxford: Oxford University Press.

Index actor 26, 35, 56–7, 68, 133 agent 58–9, 64, 127, 133 arena 61–2, 64 association, 5, 13, 73, 130 associationization 14, 83 autonomy 2, 4, 46, 68, 82 belief systems 24, 31, 38, 40 boundaries 48, 58 centralization 14, 94 change 6, 26, 40–2, 79, 136–37, 148–50 citizen 74–5, 80 company see ﬁrm company-ization 13–14, 66, 82–3, 89 consultants 40, 66, 97–9, 109, 117, 143 ‘Corporate social responsibility’ 4 constructing organizations 44, 45, 55–6 cropping-up 107, 112, 122–3 customer 74–5, 80 decentralization 14, 95 decision making 18–19, 78 decoupling 22, 106, 113 diffusion 15, 108–12, 125–6 discourse 4, 107, 112, 117–21, 123–25 expert 4, 54, 61 fashions 5, 66, 82, 97, 111–12, 124, 130, 144 ﬁrm 2, 5, 56, 73–8, 88, 130, 133 ﬂexibility 75, 94, 150–1 forgetfulness 99–100, 102 generalization 23, 79, 86, 124, 151 heterogeneity 18–20, 26, 42, 105–08, 112, 119 hierarchy 2–3, 50, 59, 142 homogeneity 18–20, 105–08, 112, 119–21 hope 9, 17, 19, Ch. 8 hypocrisy 4, 78–9, 137

ideal types 24–5, 27, 31, 73 identity 2–3, 46, 49, 58–9, 142 implementation 8–9, 95, 99, 103, 115–16, 120, 124, 128, 131–2, 145, 148 ‘incomplete’ organization 44, 61, 64 individual 26, 34–6, 89, 151 institution, institutionalized 2, 7, 9–10, 21–4, 28, 31–2, 63–4, Ch. 4, 106, 128–9, 133 legitimacy 24, 34, 36–7, 74–5 localization 23 market 10, 25–6, 28, 33, 68–9 ‘new public management’ 11–12, 54, 133 organizational types 5, 14 see also Association; Firm; State performativity 10–11 political organization 73–80, 87 politicization 13–14, 81–2, 86, 88–9 power 31, 68, 127–33 problems 4, 77–8, 84, 92–4, 102, 117, 130, 132 public organizations 55, 105–06, Ch 3 rationality 3, 15, 41, 51–3, 59, 80, 142 reformees 16, 63, 65, 134, 144, 146, 148 reformers 6, 39–40, 50, 65, 99, 121, 127–36, 138–41 responsibility 11, 31–3, 39, 41, 53, 59, 62, 67–8, 138 solutions 78, 86, 96–7, 102, 113, 117–18, 130 stability 25, 41–2, 136–38, 150 standards 5, 65, 69, 82, 106–8, 128, 130–33, 142 state 5, 39, 47, 55, 59, 68, 73, 81–2, 86, 130 talk 3, 7, 22, 79, 112–17, 135, 144–45 ‘true’ organization 2–3, 7, 12, 39, 56, 133, 141–42