Cooperation and Conflict the Nordic Way: Work, Welfare, and Institutional Change in Scandinavia 9783110444285, 9783110444278

Citation preview

Fredrik Engelstad, Anniken Hagelund (Eds.) Cooperation and Conflict the Nordic Way. Work, Welfare, and Institutional Change in Scandinavia

Fredrik Engelstad, Anniken Hagelund (Eds.)

Cooperation and Conflict the Nordic Way

Work, Welfare, and Institutional Change in Scandinavia

Managing Editor: Andrea S. Dauber Associate Editor: Dieter Bögenhold

Published by De Gruyter Open Ltd, Warsaw/Berlin Part of Walter de Gruyter GmbH, Berlin/Munich/Boston

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 license, which means that the text may be used for non-commercial purposes, provided credit is given to the author. For details go to http://creativecommons.org/licenses/by-nc-nd/3.0/.

Copyright © 2015 Fredrik Engelstad, Anniken Hagelund ISBN: 978-3-11-044427-8 e-ISBN: 978-3-11-044428-5 Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.dnb.de. Managing Editor: Andrea S. Dauber Associate Editor: Dieter Bögenhold www.degruyteropen.com Cover illustration: © Sébastien Bonaimé

Contents Preface

XII

Fredrik Engelstad, Anniken Hagelund 1 Introduction: Institutional Change in Neo-Corporatist Society 1 1.1 Scandinavian Specificities 3 1.2 Understanding Institutional Change – Theoretical Inspirations 1.3 The Politics of Compromise 10 1.4 Stability or Disintegration? 13 References 15

6

Cathie Jo Martin 17 Negotiation and the Micro-Foundations of Institutional Change Models and Determinants of Institutional Change 19 The Micro-Foundational Underpinnings of Institutional Change 22 Prototypes of Negotiation 23 Negotiation Myopia 24 Rules to Overcome Negotiation Myopia 25 Rules for Collective Engagement and the Institutions of Consensual Democracies 27 2.2.5 Rules for Collective Political Engagement and Institutional Change 28 2.3 Conclusion 31 References 33

2 2.1 2.2 2.2.1 2.2.2 2.2.3 2.2.4

PART I: Working Life Institutionalization – Negotiating Large Scale Social Change Fredrik Engelstad 36 3 Property Rights, Governance, and Power Balances 3.1 Mechanisms of Institutional Change 37 3.2 A Brief Note on Property Rights 39 3.3 The Empirical Landscape: Background 40 3.4 Case 1: Job Related Co-Determination 41 3.5 Case 2: Employee Participation in Top-Level Decision-Making 3.6 Case 3: Gender Quotas to Boards of Directors 48 3.7 Concluding on Mechanisms and Power Relations 51 References 54

45

Eivind Falkum 4

Institutionalization and Dynamic Change of Institutions – the Basic Agreement and Tripartite Structures in Norway 56 4.1 Theoretical Perspectives on Institutional Design, Development and Resilience 57 4.2 Institutionalization of Working Life 58 4.3 The First Ideas of Economic Democracy 59 4.4 Designing the Basic Agreement in 1935 61 4.5 Institutional Design of Tripartite Structures 64 4.6 Changes in Institutional Contexts 65 4.7 Industrial Democracy and Experimental Design of Workers’ Participation 67 4.8 The Change of the Basic Agreement in 1966 69 4.9 Positive and Negative Feedback of Institutional Change 70 4.10 Institutional Design and Designers 71 4.11 Path Dependencies, Resilience and Institutional Matrix 73 References 75 Inger Marie Hagen 5

Participation and Co-Determination: Why Some Arrangements Fail and 77 Others Prevail 5.1 Spheres of Democracy 79 5.2 Company Assembly and Employee Representation on the Board 81 5.2.1 Company Assembly and Corporate Governance 84 5.3 ”The Democratic Company Assembly” in 2014 85 5.3.1 Coverage 1: Company Assemblies 85 5.3.2 Coverage 2: Board members 86 5.3.3 Contested Arrangements? 88 5.4 Several Attempts – One Success? 89 5.4.1 1920 and 1945 90 5.5 Institutional Change and Consistence 92 5.5.1 Effects of Legal Changes 92 5.5.2 Layering 93 References 95 Mari Teigen 6 6.1 6.2 6.3

The Making of Gender Quotas for Corporate Boards in Norway Theoretical Perspectives 97 A Note on the Data 98 National Preconditions and Processes 99

96

The Norwegian Gender Equality Policy Institution 101 Positive Action and Gender Quota Arrangements in Norway 101 Employee Representation on Corporate Boards 103 Deregulation Meets the Institutionalization of Gender Equality Policies 104 6.3.5 Women in Management 106 6.3.6 The Public Debate on Gender Quotas for Corporate Boards 106 6.3.7 The Policy Agency Context 109 6.4 Discussion 111 6.5 Conclusion 113 References 115

6.3.1 6.3.2 6.3.3 6.3.4

Cathrine Holst 7

Institutional Variation and Normative Theory: Lessons from a Local Equal Pay Controversy 118 7.1 Is Equal Pay Justified? The Standard Academic Debate on Comparable Worth 119 7.1.1 Standard Contra Arguments 119 7.1.2 Standard Pro Arguments 121 7.2 A Local Controversy: Equal Pay in a Scandinavian Context 123 7.2.1 Additional Contra Concerns and Arguments 124 7.2.2 Additional Pro Concerns and Arguments 126 7.3 Why is the Local Controversy Seemingly Richer? 128 7.4 Lessons for Normative Theory 131 References 132

PART II: Institutional Change in Work and Vocational Education Dag Olberg 8 Regulating the Temporary Layoff Institution – Coalitions and Drift 8.1 Introduction 136 8.1.1 Chapter Outline 137 8.1.2 Research Questions 137 8.1.3 Data 138 8.2 The Crisis – Overcapacity, Working Time and Employment Relations 8.2.1 European Comparative Institutional Perspectives 139 8.2.2 Nordic Comparison 140 8.2.3 The Norwegian Temporary Layoff Institution 142 8.3 Regulatory Changes – Variations According to Labour Market Situation 143

136

139

8.3.1 The Ebb and Flow of Regulatory Changes 143 8.3.2 Regulations 2009 – The Role of the Social Partners 144 8.3.3 The Extent of Temporary Layoffs 145 8.3.4 Regulations and Re-Regulation – 2012 146 8.3.5 Change in Practices – Unemployment Benefits and Temporary Layoffs 147 8.3.6 The Accord 148 8.4 The Temporary Layoff Institution and the Role of the Social Partners 149 8.4.1 Political Coalitions – Bipartite Coalitions Within the Tripartite Model 149 8.4.2 Interest Construction – Mobilization and Power, Renegotiations 150 8.5 Conclusion 153 References 155

Jørgen Svalund 9

Cooperation and Power in Labour Adjustment Choices: 157 A Nordic Perspective 9.1 Analytical Framework 159 9.1.1 Institutions, Agency and Outcomes 159 9.1.2 Labour Adjustment Strategies in the Face of Institutional Regulations 159 9.1.3 Power and Compromises Within Long-Term Reciprocal Relationships 160 9.2 Data 161 9.3 Labour Adjustments in the Nordic Countries: Institutional Characteristics and Alternatives 162 9.3.1 Dismissal Regulation 162 9.3.2 Regulation of Temporary Layoffs 163 9.3.3 Unemployment Benefits and Early Retirement Systems 163 9.4 Power, Cooperation and Institutions within Three Industries in Norway 164 9.4.1 Manufacturing 165 9.4.2 Construction 166 9.4.3 Newspapers 168 9.5 Power, Cooperation and Institutions Within Manufacturing in the Nordic Countries 169 9.5.1 Denmark: Dismissals as an Employer Prerogative – Employer Choices without Union Interference 170 9.5.2 Finland: High Level of Employer Discretion during Dismissals Combined with Limited Veto Possibilities 171 9.5.3 Sweden: Lack of Temporary Layoff Mechanism Combined with Strict Dismissal Rules Laid Ground for Union Power 172 9.6 Discussion and Conclusion 173 References 177

Heidi Nicolaisen, Sissel C. Trygstad 10

Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market 179 10.1 Dualization of Working Conditions and Labour Relations? 180 10.1.1 Methods and Data 181 10.2 Analytical Perspectives on the Prospects of Change 182 10.3 The Foundation of the Norwegian Industrial Relations Model 184 10.3.1 The Cleaning Industry – on the Margins of the Norwegian Model of Labour Relations 185 10.3.2 The Structure 185 10.3.3 The Workforce 185 10.3.4 The Labour Relations 186 10.3.5 The Competition 186 10.3.6 Different Degrees of Decency 187 10.4 Measures to Prevent Dualization 188 10.4.1 Extension of Collective Agreements 189 10.4.2 Approvals and ID Cards 189 10.4.3 A Mix of Soft and Hard Regulations 190 10.5 The Introduction of Hard Regulations – A Window of Opportunity 190 10.5.1 The Change Capacity of Hard Regulations – The First Experiences 192 10.5.2 Something Has Already Worked 194 10.6 Discussion and Concluding Remarks 195 References 199 Torgeir Nyen, Anna Hagen Tønder 201 Cooperation and Reform in Vocational Education and Training Historical Background: A Strong Coalition Between Employers in Crafts and Industry 204 11.1.1 Apprenticeship Training in Industry Regulated in Collective Agreements 204 11.1.2 The Post-War Period: The State Takes a More Active Role 206 11.1.3 The 1980s: Revitalization of Apprenticeship Training 208 11.1.4 Reform 94: Apprenticeship Training an Integral Part of Upper Secondary Education 209 11.1.5 Reform 06: Broader Courses with Opportunities for Specialization 211 11.1.6 Recent Developments 211 11.1.7 Institutional Changes 213 References 216 11 11.1

PART III: Negotiations in Welfare State Institutions Anniken Hagelund, Axel West Pedersen 12

To Reform or Not to Reform? Explaining the Coexistence of Successful Pension Reform and Sick Pay Inertia in Norway 220 12.1 Introduction 220 12.2 Theoretical Perspectives on Social Policy Reform 221 12.3 Case I: Sickness Insurance 223 12.4 Case II: Pension Reform 225 12.5 Actor Constellations and Strategic Games 227 12.6 Institutional Dynamics and Path Dependency 231 12.7 Discourse and Framing 233 12.8 Conclusion 236 References 240 Mia Vabø 242 13 Changing Welfare Institutions as Sites of Contestation 13.1 The Contested Character of Welfare Service Institutions 243 13.2 Institutions as Constituted by and Changed by Discourse 244 13.3 The Institutional History of Norwegian Home Care 247 13.3.1 1960s and 70s – An Era of Expansion 247 13.3.2 1985-1995 – An Era of Decentralization and Democratization 247 13.3.3 1995: The Quest for Transparency and Enforceable Rights for Citizens 248 13.4 The Role of NPM as a Driver of Change 249 13.5 Processes of Change at the Operational Level 250 13.6 Self-Regulated Care Teams Facing the Purchaser–Provider Model 251 13.7 The Purchaser-Provider Organization as a Tool for Taylorization 254 13.8 How Responsible Purchasing Paved the Way for Professional Power 255 13.9 The Dynamic of Change in Social Service Institutions 257 References 259 Anne Lise Ellingsæter 14 14.1 14.2 14.3 14.4 14.5

Making, Unmaking and Remaking: The Evolution of Nordic Cash for 262 Childcare Schemes Political Parties and the Welfare State – Changing Relations 263 Incremental Institutional Change – Across Nations and Time 264 Value Competition: Equality Versus Choice 266 Current Nordic Cash for Childcare Schemes 266 Finland: Layering 268

14.6 Norway: Layering, Partial Displacement, Announced Re-Layering 270 14.7 Sweden: Layering, Displacement, Re-Layering 273 14.8 Nordic Cash for Childcare Schemes: Stability and Instability 274 References 278

PART IV: Afterword Fredrik Engelstad 15 Conflict, Compromise, Cooperation – Concluding Reflections 15.1 Flexibility – Balancing Conflict and Contract 282 15.2 Emergence and Expansion of the Tripartism 283 15.3 Reform as Routine 285 15.4 Norms in Institutional Change 286 15.5 Sustainability of Egalitarian Capitalism? 288 References 292 293

About the Authors List of Figures List of Tables Index

296

294 295

282

Preface The study of institutional change has been among the most innovative fields in the social sciences during the last decade. One obvious reason is the occurrence of empirical work of exceptionally high quality in political economy and welfare states. Simultaneously pertinent studies in social theory have appeared, highlighting a variety of mechanisms and processes of social change. Furthermore, these contributions point to new ways of conceptualizing what was conceived as the main topic of sociology by such diverse thinkers as Simmel and Durkheim – the idea of “society” as such, and more specifically the concept of modern society. The classics’ preoccupation with the society as such has to a large extent disappeared from the social sciences, and partly for good reasons; in its pure form it comes too close to a Theory of Everything. If social totality remains elusive, it may still be grasped partially as constellations of a plurality of institutions. They are the stuff societies are made of. On a lower level of generality, institutions may be analyzed both empirically and conceptually, without sacrificing theoretical rigor. Thereby such analyses may also deliver input to reflections on basic traits of specific societies. The ambition of the present volume is to continue work along these lines by collating a broad set of institutional features in one type of society, namely the Scandinavian welfare state. Albeit similar on a large set of indices, Scandinavian societies also differ in interesting ways. Up to now empirical studies with institutional perspectives have mostly focused on Sweden and Denmark; here Norway is placed at the center, however with several sidelong glances to the other Nordic countries. The present volume on working life and welfare state is the first in a series of three books. The next volume will cover changes in the public sphere, with main emphasis on effects of the new social media. The third volume will discuss democracy and institutional change, with focus on institutionalization of civil and social rights outside the specifically political realm, such as civil society organizations, education, civil service, and gender relations. These three volumes are the outgrowth out of a network project funded by the Norwegian Research Council for the years 2013-2016. Three partners have taken on practical responsibilities for the project: Department of Sociology and Human Geography at the University of Oslo, Institute for Social Research, and Fafo Institute for Labour and Social Research. Thanks to the Norwegian Research Council and to the collaborating institutes for their support to the project. And special thanks to professors Kathleen Thelen, Cathie Jo Martin and Howard Gospel for inspiring contributions to the initial workshop that brought about the present volume. Oslo, April 2015 Fredrik Engelstad Anniken Hagelund

Fredrik Engelstad, Anniken Hagelund

1 Introduction: Institutional Change in NeoCorporatist Society The Nordic model of neo-corporatist society attracts attention, not unexpectedly, in a mix of applause and disbelief. Numerous international rankings of quality of life, gender equality, welfare state provisions and even protection of property rights consistently put the Nordic countries in the top 10 or even top five groups among the world’s nations. Remarkably, The Economist (2013) followed up in a lead article: “The Nordics cluster at the top of the league tables in anything from economic competitiveness to social health and happiness”. But is this not too good to be true? Denouncements of the Nordics are not lacking. The extensive welfare state entails unacceptable taxation and corresponding losses in economic growth, according to the Cato Institute (Mitchell, 2007). Behind the curtain of the almost nearly perfect people, paternalism and cultural isolationism is reigning, Booth (2014) laments in a joking tone, seemingly with a serious intention. Now, desirable states obviously come at some cost, as it is with Nordic societies. A more noteworthy question, maybe the most challenging, is raised by a report from the think tank Civita: “Can we afford the future?” Is the welfare state as we know it a viable project? (Fasting, 2014) Reflections on the future of the Nordic model presuppose knowledge about its developments and changes over a long time period. The viability of social structures and social arrangements is dependent not only on economic resources but just as much on the fabric of institutions where they are inscribed. In studies of the Nordic model, however, from Esping-Andersen’s Politics Against Markets (1985) to the recent The Nordic Model of Social Democracy (Brandal, Bratberg & Thorsen, 2013), a broad focus on institutional change has been virtually absent. Hence, in this book processes of and capacities for institutional change are scrutinized. If the Nordics have maintained their specific models for labour market coordination and welfare state arrangements through their ability to adapt to new conditions and change institutions without losing their basic foundations, how did the institutions that enabled such patterns of incremental change emerge, and how do they work today? Compromise and coalitions are at the heart of institutional change processes in these societies. In some of the following chapters, important institutions are put in place as a result of antagonistic parties agreeing on an alternative that is the “least worst” (Falkum on the Basic Agreement or Hagelund & Pedersen on the Agreement for an Inclusive Working Life). The layering of a new set of rules on top of existing institutions seems to be a typical avenue of reform, where diverse interests are able to find common ground. But beneath compromise is conflict, even in these relatively peaceful societies. While corporatist modes of collective engagement have structured, and at times worked to mellow, processes of institutional change in working life and © 2015 Fredrik Engelstad, Anniken Hagelund This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

2 

 Introduction: Institutional Change in Neo-Corporatist Society

parts of welfare policy, there are also policy fields where the conflicts are less contained. In family policy, for example, fronts are sharper and the outright displacement of rules complement the otherwise dominant pattern of layering (Ellingsæter, this volume). The scholarly literature of the past decade has seen a fruitful growth, both theoretically and empirically, in theories of institutional change. The theoretical development opens up a fruitful avenue to empirical studies of social change on a large scale. This is what we offer in this book: Empirically grounded analyses of institutional change in core working life and welfare state institutions in Scandinavia, with a special focus on Norway, ranging from property rights, boardroom politics and wage formation to old-age pensions, care work and childcare policies. There are strong reasons for skepticism over general theories of social change comprising societies as a whole. What is usually called “societies” in the meaning of nation states are too complex and too fuzzy to be the object of all-comprising theories. In various ways, this was indicated a long time ago by Raymond Boudon (1985) in his thorough study of various theories of social change and by Michael Mann (1986) conceiving large-scale social change as taking place in several semi-autonomous sets of social networks. This does not preclude large-scale analyses of social change, but requires disaggregation of processes to a somewhat lower level – that of social institutions. The strength of recent theories of institutional change is that they make credible the ambition to combine broad institutional level theory with thorough and rather detailed empirical studies. An early example is the collection on Varieties of Capitalism (Hall & Soskice, 2001), where the concept of “bundles of institutions” plays a major role. A weakness of this fruitful work is that it tends to be static, assuming that institutions are virtually immovable. Recent work covering parts of the Nordic model, overcoming this weakness, is the comparative research by Kathleen Thelen on Varieties of Liberalization (2014) and by Cathie Jo Martin and Duane Swank on The Construction of Business Interests (2012). Theories of institutional change serve as stepping stones for the following empirical studies. In describing the emergence of central parts of the model, a key point is to call attention to the inconsistencies, incompatibilities and conflicts that necessarily are part of any social regime, including the Nordic model. The conception of a Nordic model does not assume the Scandinavian countries to be similar in most or all matters of interest. There is considerable variation in political regimes and economic structures (Sejersted, 2001; Stråth, 2001), and recent changes have to some extent pointed in different directions. Thereby, comparison between countries in the Nordic area is of interest – something that is found in several of the contributions here. At the same time, the main focus is on the case of Norway. One methodological advantage of this approach is that Norway is the country where neo-corporatism has developed the furthest and has been best preserved. Hence, it represents one extreme value on a general dimension of “social pacts”, or state/society interplay, which to varying degrees is found in a number of European countries (Avdadic, Rhodes &



Scandinavian Specificities 

 3

Visser, 2011). Drawing up a dimension of degrees of neo-corporatism in Europe, the UK would occupy one extreme position of low institutionalization of employment relations, whereas Norway would be placed at the other end of the spectrum. As an extreme case, Norway is well suited to study institutional changes which are not primarily driven by crisis, but where social and political reform has the character more of an ongoing concern.

1.1 Scandinavian Specificities The institutional specificities of Scandinavian societies date back much further than the advent of social democracy. By the beginning of the 18th century, they had a relatively efficient state bureaucracy. The existence of a formalized Protestant state church also contributed to the formation of the centralized state power. Unlike several other European societies, the combination of secular and religious power was never seriously contested. This also had a crucial consequence for later developments of democracy. As part of the preparation for the compulsory confirmation of the Protestant Church, adepts had to learn to read the Bible in the national language. Thus, literacy in the population was almost complete by the beginning of the 19th century. In turn, this became a precondition for diffusion of liberal thinking during the industrialization and modernization processes of the mid-19th century, mostly in Norway and Denmark, not least in opposition to monarchic power and its central bureaucracy. Thereby, the processes of democratization, stretching over the entire 19th century, took the shape of democratization from the inside, within the confines of a relatively well-functioning state. Reform, not revolution, became the obvious and legitimate mode of social change. In the long run, class conflict, in line with Dahrendorf’s (1959) analysis, was transformed into an institutionalized class compromise. Social movements came to play a key role in Scandinavian political history in the latter half of the 19th century and for Norway were most pregnantly formulated by Stein Rokkan (1967) as the “counter-cultures” of the temperance movement, the language movement, and the lay Christian movement. In Denmark, the movement founded by the liberal preacher Grundtvig had a similar influence on education and cultural life. A common trait in these movements was their politically transformative power; they established new policy fields and politically based institutions by “conquering” a decisive position in the state within their respective fields. The most powerful transformation, however, stemmed from the labour movement. Almost simultaneously, around the turn of the century, scattered trade unions were united into a trade union federation in Sweden, Denmark and Norway. In parallel, employers organized into national federations as well. The road was thereby paved for national wage agreements between the labour market parties and subsequently for the establishment of basic agreements specifying modes of collective bargaining

4 

 Introduction: Institutional Change in Neo-Corporatist Society

and formalizing rights and duties on both sides of the employment relation. In Denmark, the first basic agreement was signed in 1899, Norway and Sweden followed in the latter half of the 1930s. In all three countries, the emergence of organizing labour market relations was followed up in politics by political regulations of labour conflict through the establishment of labour courts and legislation regulating strikes and lockouts. The field of compromise and cooperation grew from the early post-war period, resulting in broadly based national wage bargaining systems. It eventually also led up to the parallel introduction of new forms of workplace democracy in the 1970s in all three countries, embracing employee participation in health and safety committees on the shop floor, as well as employee representation on the boards of joint-stock companies. Together they constitute a neo-corporatist system, with precarious power balances between the state, employer federations and trade union federations. The willingness to compromise on the opposing labour market interests was not only a result of internal organizational development, but should be seen in light of certain characteristics of the economy. Despite different economic resource bases, the Scandinavian countries have in common that they are small and open economies; the fierce competition in the world market invites domestic cooperation and simultaneously a high degree of attention to the necessity of increasing productivity (Katzenstein, 1985; Moene & Wallerstein, 2003). The class compromises of the early 1900s in turn became the basis for the growth of the welfare state from the 1940s, most clearly pronounced in the Swedish concept of “folkhemmet” – the home of the people. This also secured the trade union movement a key position not only in bargaining over wages and working conditions, but in the much broader fields of social policy, pension systems and vocational training. A marked expansion in the Scandinavian neo-corporatist system up to the 1980s was followed by considerable adjustments in the early 1990s and was most clearly pronounced in Sweden, when employers withdrew from any form of direct tripartite cooperation. National wage bargaining has to some extent been decentralized, more so in Sweden and Denmark than in Norway. At present, Norway has by far the most formalized bargaining system of the three, as well as the strongest tripartite elements. In their agreement on the “Solidarity Alternative” of the 1990s, employers, employees and the government collaborated towards the aims of competitiveness and full employment through central coordination and moderate wage growth. The spirit continued through a row of public commissions with the social partners present, that work to update and adjust the parties’ shared understandings of problems and secure continued support for the model of coordination (Andersen, Dølvik & Ibsen, 2014). Modifications and changes in Nordic neo-corporatism should not detract from the fact that contact and cooperation between unions, employer organizations and the state is still closer in all of Scandinavia than in the rest of Europe (Avdagic et  al., 2011). Cathie Jo Martin (this volume) stresses precisely this point – the development



Scandinavian Specificities 

 5

of institutions for political negotiation and rules for political engagement as a fundamental characteristic of the Scandinavian countries. Adjustments have also been made in welfare state provisions. In Sweden and Denmark, there are clear signs of retrenchment, while the generosity of Norwegian social security schemes to a greater extent has been maintained. However, in all countries, the social political tendency goes in the direction of activation, and the high level of decommodification stressed by Esping-Andersen (1990) is being moderated. Active policies are established with the aim of incorporating ever new groups – single parents, refugees, those of ill health, the elderly – into the labour market. Old-age pension schemes have been the object of renegotiations and significant alterations both in Norway and Sweden, and further adjustments seem necessary and highly probable as the population is ageing despite comparatively high fertility rates (Hagelund & Pedersen, this volume). Globalization means globalization of capital flows and a relative weakening of political governance in the economy. In particular, global migration flows imply internationalization of labour markets, in some industries putting considerable pressure on established regulations in working life. Sweden has long been a country of immigration and an exceptional case in Europe with respect to its liberal policies towards refugees. In Norway in particular, the extension of the European Union (and thus of the European Economic Area of which non-EU Norway is a member) combined with high economic growth and demand for labour have led to large-scale immigration from the new EU countries, thereby transforming the basic premises of working life arrangements in industries such as construction and cleaning (Nicolaisen & Trygstad, this volume). Nevertheless, adjustments have made it possible to retain the main elements in the Nordic model (Dølvik, 2007; Thelen, 2014). Challenges to the model are notable but may possibly be overcome as long as popular support for the welfare state is unshaken, as it seems to be for the foreseeable future. The Scandinavian societies – and Norway to a possibly higher degree – have moved from a mode of change connected to building the welfare state, to a mode of defence and adjustment of the model in the face of challenges from outside. This gives the Nordic countries a particular reform dynamism based in large-scale bargaining and compromises. So far, emphasis has been placed on the social partners – the labour market organizations – and their relationship with the state. But we cannot consider the state without taking into account the significance of political parties and changing government power. Traditional theory of partisan politics assumes a linear and direct relationship between the type of party in power and policy output – for example, the more left the government, the more generous welfare policies will result. Recent research questions this assumption, underscoring the importance of context and institutions to understand the preferences and actions of governments (Häusermann, Picot & Geering, 2012). In the Nordic context, the power resource approaches of Walter Korpi (1983) and Gøsta Esping-Andersen (1985) have emphasized the role of

6 

 Introduction: Institutional Change in Neo-Corporatist Society

social democracy in shaping core institutions in ways that have also engendered vital coalitions between the working and middle classes. The Scandinavian countries are undoubtedly characterized by stability and consensus over crucial policy areas. This is reinforced by the dominance of egalitarian values. Still, the party system is also a source of tension and political conflict that may not result in grand upheavals when government power shifts, but instead, as each party tries to put its mark on policy without rocking the main course of the Nordic model, leads to the myriad of incremental changes and adaptations that make up the stability through the change patterns we are exploring here. In the following chapters, cases of this reform dynamism, with its pit holes, backlashes and processes of trial-and-error, are analyzed in an institutional perspective to make visible the balance of stability and change, and thereby to contribute to reflection over the conditions of sustainability of the Nordic model generally, and more specifically of the Norwegian variant.

1.2 Understanding Institutional Change – Theoretical Inspirations Like many other social science concepts, the notion of “institution” is a slippery one. In this book, it is conceived quite conventionally as macro regulations of social behaviour by organizations, groups and individuals. Institutions are based on formal elements such as legislation, bureaucratic regulations or large-scale agreements. Simultaneously, rules and regulations must be interpreted and are the object of normative reflection. Hence, normative and cultural-cognitive aspects are equally part of institutions, as pointed out by Scott (2008, p. 48) and Schmidt (2002). Institutions vary in their scope and often take on a nested character; as a simple example reflected in several of the chapters of this book, working life may be seen as an institution, in a more narrow meaning the enterprise is an institution, and within the enterprise, the board of directors or employer–employee relations constitute institutions. In the political sphere, it also makes sense to regard long-term policy engagements as institutions (Pierson, 2006; Teigen, this volume). But why concentrate on institutions if the topic is social change? Alternative modes of thinking are not lacking. Typical historical expositions mainly rest on a narrative approach. An example close at hand is Francis Sejersted’s (2011) masterly dual portrait of Sweden and Norway in the 20th century, conceived in terms of the rise and fall of social democracy. At the other end of the spectrum, institutions are deprived of their specific meaning by being reducible to rational choices, as in the seminal discussion by Demsetz (1967) of the transformation of commons into private property. Important as they are, these approaches too easily become formulaic. Politics are enacted in time, Paul Pierson (2004) emphasized, and develops along a variety of paths. Historical transformation must take in both regulatory and innovative perspectives, Kathleen Thelen (2012) underscores. The advantage of the focus on institutional



Understanding Institutional Change – Theoretical Inspirations 

 7

change is that complex patterns of social action are highlighted on the one hand by depicting social behaviour as relatively stable and rule governed, but on the other hand as directed to changes in these rules, be it by specific reformative actions or by aggregate effects of a plethora of actions and preferences. That regulations are both restricting and enabling is only half of the story; they are also affected by dynamic processes changing the patterns of restriction and enablement. This is true for the distribution of social resources; in parallel, institutional variation becomes a source of variation in normative deliberation (Holst, this volume). Approaches to institutional change are manifold (Hall & Taylor, 1996; Thelen, 1999, 2012). The aim here is only to present the conceptions which are most relevant to the empirical studies collected in the book. A relative simple model, which stands in contrast to many (but not all) of the studies in the book, is characterized by the notion of “punctuated equilibrium”. This implies that societies mostly endure in a sort of relative equilibrium and proceed along existing paths. Changes are due to external shocks, be it a war or an economic crisis, which constitutes a critical juncture. A “window of opportunity” is opened for the creation of new rules and a new course of action, until the occurrence of the next external shock with ensuing revision of rules and policy paths (Collier & Collier, 1991). Albeit simplistic, this theory has an intuitive appeal. In many instances, it takes some sort of crisis consciousness before reforms are initiated, such as, for example, in the legal regulation of banks and financial markets occurring more easily in the aftermath of a financial crisis. Among the empirical studies, critical junctures are depicted in the chapter by Nicolaisen & Trygstad on the effects of migration waves to the Norwegian labour markets after the extension of the EU in the early 2000s. Likewise, the introduction of New Public Management in the home care sector became a critical juncture for trade unions and care workers alike, as discussed by Vabø (this volume). A problem with the theory of critical juncture lies in its conception of the repetitiveness of regular social processes. Shifting the focus to incremental transformation opens up a more nuanced picture. In her contribution to this volume, Cathie Jo Martin goes into the micro foundations of institutional change and explores various modes of negotiation and how they affect outcomes. Central in this connection is how negotiating parties define their own roles and long-term prospects. A salient factor is the degree of myopia in bargaining, the tendency for the parties to focus solely on short-term gains and their own narrowly-defined interests, at the cost of more advantageous outcomes in the long run. Myopia is diminished by reciprocal trust and by institutional structures, such as repeated interactions and the use of non-partisan expertise – features that are crucially present in, for example, collective wage bargaining at the national level. In his Politics in Time (2004), Paul Pierson refines the theory of critical junctures and path dependency in order to explain tendencies towards institutional stability and not simply take them for granted. Here, he points to four types of mechanisms. First is problems of coordination. In line with theory of collective action, even if actors

8 

 Introduction: Institutional Change in Neo-Corporatist Society

have a common interest in a given goal or in avoiding a certain outcome, deficient patterns of contact or general preferences for acting as a free rider result in the preservation of status quo. Second is processes of positive feedback. Institutional arrangements working fairly well over a certain period of time tend to be self reinforcing. By accepting a given set of rules, actors also adjust to them, maybe even broaden their extension. This in turn affects attitudes and expectations, thus narrowing the relevant set of alternatives in future dispositions. The reinforcement and extension of the Basic Agreement in Norway in the post-war years is a case in point (Falkum, this volume). Third is veto points. Actors may exercise control over salient social resources and thus be able to prevent changes, even if they are deemed rational and necessary by other actors. Unsuccessful attacks on sick leave payment provide a telling illustration (Hagelund & Pedersen, this volume). Fourth is asset specificity. The resources controlled by actors, such as money, real estate or technological knowhow, are limited to a specific set of situations and transactions. Given that changes also affect the value and usefulness of resources, powerful actors will have a strong motive to preserve arrangements in which they get the most out of their own resources (Pierson, 2004, pp. 142-153). These are powerful mechanisms contributing to the continuation of a given institutional pattern. Moreover, what we see in our cases are situations of relative normality and certainly no real crisis or shock where changes are still occurring. There were no external shocks setting in motion cash for care policies (Ellingsæter, this volume) or women’s boardroom quotas (Teigen, this volume), but rather changing political constellations or even single-handed political entrepreneurs. However, Pierson also points to sources of institutional change, and underscores that such changes may be invisible in the short run and need to be observed over a long time period. One source is changes in the environment, much in line with the theory of external shocks. A second set of mechanisms has to do with changes in preferences. Actors change their goals or their conception of the means to reach these goals. There is necessarily turnover among the incumbents of social positions, and new people often come up with new ideas. The loss of confidence among trade union officers in the Company Assembly as a democratic instrument is a good example (Hagen, this volume). Finally, institutions have multiple effects, whereas actors are myopic, with the unavoidable result of important sets of unintended consequences (Pierson, 2004, pp. 109-122) and unpredictable processes. The complex process leading up to the adoption of gender quotas to boards of directors (Teigen, this volume) is a case in point. The work of Kathleen Thelen marks a contrast to the idea of critical junctures and path dependency. She puts a different and even stronger emphasis than Pierson on the effects of permanently ongoing actions, which, viewed in isolation, may appear insignificant but nevertheless may entail significant aggregate effects. Institutions are changed not only due to political decisions, but are also shaped by ongoing processes both outside and inside the institution. Thelen’s emphasis on



Understanding Institutional Change – Theoretical Inspirations 

 9

agency accords great importance to power and power struggles as explanations of the emergence and change of institutions. At the same time, she underscores the salience of relative stability of institutional patterns as a context for action. Positioning herself in relationship to the Varieties of capitalism literature, she criticizes the tendency to exaggerate the stability in the institutional bundles, whereas in opposition to what she terms Pragmatic Constructivism, she underscores the salience of structural restrictions to political action (Thelen, 2012). In a series of papers and chapters, Thelen and colleagues have elaborated the mechanisms of institutional change through aggregate processes (e.g. Thelen, 2004; Streeck & Thelen, 2005; Mahoney & Thelen, 2010). Four types of mechanisms are discussed: displacement, layering, drift and conversion. Layering is probably the most common type of reform; here, a process of sedimentation is taking place by adding new elements to the old ones. An example is the introduction of New Public Management in public services while existing modes of governance were preserved (Vabø, this volume). Displacement is a more radical form of change, where one institutional arrangement is supplanted by another. A possible illustration is the transfer of the main responsibility for labour protection from an external control agency to a committee manned by employers and employees within the company (Engelstad, this volume). Displacement and layering are processes that to a large extent are linked to large-scale political decisions. Drift, on the other hand, is the result of changes in the environment leading to changes in the functioning of the institution. Thelen’s work on changes in vocational training in Germany is a telling example; similar processes are depicted in the emergence of the present system of vocational training in Norway (Nyen & Tønder, this volume). Another example of drift is found in Olberg’s analysis of changes in the institution of temporary layoffs. Finally, conversion means that an institution remains formally unchanged but fulfils quite different functions than those performed earlier. Svalund’s (this volume) analysis of dismissal processes in Scandinavian enterprises exemplifies how institutional variation appears when rules are interpreted and enacted in different ways locally despite unchanged formal rules. The Norwegian Labour Inspection Authority may offer another example. The institution of labour inspection has remained virtually untouched over a long period, but its working has changed drastically. Physical control by visits to companies was replaced by requirements of companies to set up health and security plans; it is these documents that are the object of inspection, not the physical workplaces. However, when the new regime of compliance and trust is threatened in some industries, the possibility of re-conversion is introduced (Nicolaisen & Trygstad, this volume). Ideological factors and the power of ideas and discourse are aspects that are seemingly neglected in much of the institutionalist literature, a lacuna which Vivien Schmidt (2002, 2009) among others has attempted to bridge with the headline of discursive institutionalism. She has pointed to how institutional change is facilitated or barred by the discourses policy makers and politicians use to negotiate and argue for policy reform. Schmidt distinguishes between four fundamental functions performed

10 

 Introduction: Institutional Change in Neo-Corporatist Society

by discourse in the policy process. The cognitive function of discourse is to define the purposes of policy reform and the problems it is set to solve and to offer policy instruments and appropriate methods. This is illustrated by Hagelund & Pedersen (this volume). The normative function of discourse involves demonstrating the policy programme’s consistency in terms of political goals and ideals – for example, by appealing to national values and identities. A related topic is brought up by Holst (this volume) in her discussion of normative variation in conceptions of equal pay. While these two functions belong to the ideational dimension of discourse, the next two make up the interactive dimension. Here, the policy discourse plays a coordinative function by creating a common language (epistemic community, discourse coalition) whereby different actors can communicate and come to agreement on a specific policy programme (e.g. Ellingsæter on care for cash policies). Finally, the communicative function of discourse is about how policy actors are able to communicate a policy programme to the public (e.g. Teigen on the introduction of gender quotas on boards.) There can be no interesting Theory of Everything; this is true also for theories of institutional change. Each of the theories presented above have their strong and weak sides. Although partly formulated in a language of critiquing the other, there are no serious inconsistencies between them concerning methodological or epistemological perspectives. They all are based on a general conception of agency, even though they to various degrees specify structural elements or processes of social action. The main difference rests in their varied focus on process types, dominant mechanisms and time perspectives. The fruitfulness of combining different theories follows from the structure of the problem to be studied.

1.3 The Politics of Compromise The empirical studies in this volume to a large degree examine reform processes which are not so much the product of external constraints, but are born out of general social reformism. In most cases, efficient compromises cannot be enforced by one party on the basis of superior strength. It takes bargaining resulting in some sort of agreement to secure stability. A salient point is whether bargaining takes place solely on the basis of the parties’ unilateral interests mediated by their reciprocal dependency, or whether it is linked to institutionalization processes. That bargaining will be more stable if it takes place within the context of an already legitimate institutional framework, and even more so if it leads up to further processes of institutionalization, had already been pointed out by Dahrendorf (1959). This will diminish tendencies of negotiation myopias and avoid excessive attention to short-term gains and corresponding self-serving motivation, as demonstrated by Martin (this volume). She suggests that the inclusion of non-partisan expertise into the negotiation process, repeated interactions and penalty defaults are institutional factors that work to encourage such inclusive negotiation. These are the patterns of institutionalized



The Politics of Compromise 

 11

bargaining that tend to characterize much of Scandinavian working life and welfare states. What the different cases are unravelling, however, are the myriad of ways in which these politics of compromise are enacted. A historically crucial example is the establishment of basic agreements in all the Scandinavian countries. Their emergence presupposes the solution of a collective action problem, as Falkum’s chapter illustrates. It was the complex constellation of state, employer association and trade union federation that made these agreements possible. The labour market parties made ideological sacrifices in order to reach a compromise but won a position of political influence, as well as labour peace. In Norway, on a par with the other Scandinavian societies, the wide-reaching class compromise established with the Basic Agreement paved the way for tripartite wage bargaining and became a decisive building block for the welfare state – both products of the 1960s. The reforms of workplace democracy of the 1970s were in many ways traceable to quite strong tensions in working life during the 1960s, but they also reflected an ongoing reform endeavour driven by the Labour Party and the trade union federation, and to a large extent accepted by the employers as well. The character of trial-and-error in institutional change is highlighted in the long-term development of workplace democracy in the persistence of employee representation on the board of directors and the demise of the Company Assembly (Hagen, this volume). The drive behind these institutional changes is not only material interest and peace in the workplace, but normative commitment as well. Albeit not reducible to material interests, these norms are to a large extent shaped by the institutional frameworks of the class compromise and were not feasible without a dialogue between the three parties of state, labour and capital. This is demonstrated by Holst (this volume), showing how alternative conceptions of justice in wage formation are molded by the organizational and institutional context – in this case, the order of collective bargaining. The same is true for variation in conceptions of property rights, as discussed by Engelstad (this volume). Property rights remained unmentioned in the Basic Agreement, something that was a condition for the establishment of the original agreement in 1935, and simultaneously it allowed for redefinitions of the precise meaning of property rights in the light of other, competing normative concerns. Scandinavian neo-corporatism and structures of tripartite negotiation also include the labour market parties in wide-ranging bargaining on matters that point beyond the labour market in a narrow sense. Publicly funded pension systems and sick leave payment are obvious examples (Hagelund & Pedersen, this volume). The existing tripartite bargaining arrangements, or what remains of them, become a natural framework for such bargaining processes. Regarding sick pay, employers partly fund the sick pay (for the first two weeks), and the workplace has been identified as the crucial arena for measures to reduce sickness absence (through improving the work environment and adapting work conditions to workers with health problems). In regards to old-age pension, reforms of publicly funded pension schemes will interact

12 

 Introduction: Institutional Change in Neo-Corporatist Society

with the structure of labour market-based pension schemes, making reforms of the latter crucial to the implementation of the targets of the former. Most of the time, reforms modify but still build upon existing arrangements. Only in exceptional cases are they implemented from scratch. This highlights the salience of layering as a mode of institutional change in the Scandinavian reform societies, as in the generous Norwegian sick pay where new layers of control and activation measures were added to the economic instruments. The result of layering may be that the old and new elements reinforce each other – but also the opposite, resulting in long-term tensions between the old and new parts. Over time, the old elements may contribute to weaken the new, thus preserving the status quo, or the new parts may eventually displace the old ones. In this book we find many such examples of tensions between elements in reform. One is the failure of the Company Assembly (Hagen, this volume). Another is the conflict between New Public Management (NPM) governance and professional norms in local welfare services (Vabø, this volume). Still another illustration emerges from the analysis of reforms of the Norwegian system of vocational education and training (VET). Stronger elements of theoretical training have been introduced into the VET system. One consequence is signs of academic drift in the vocational training system, especially in service educations. Many students leave the VET programme and transfer to a higher education path, thus in practice undermining the ambitions of strengthening VET by way of giving such training a more solid theoretical basis (Nyen & Tønder, this volume). Several chapters in this book point towards the driving force of ideas and discursive processes for institutional change. Anne Lise Ellingsæter identifies a powerful discourse of “parental choice” in the evolvement of Scandinavian family policy. The cash for childcare schemes that have emerged in both Finland, Norway and, to a lesser degree, Sweden over the past few decades stand in peculiar contrast to the otherwise strong emphasis on women’s employment and subsidized daycare for children below school age. The ideas date back to the 1960s male breadwinner/ housewife family ideals. Over time the original arguments have been adapted to the social realities and political rhetoric of the post-industrial society, where housewife ideals are rejected but where ideas about parental choice with respect to childcare have attracted considerable support. But cash for care remains a controversial policy, and the choice rhetoric divides as much as it unites. Hagelund & Pedersen (this volume) illustrate another type of discursive dynamic, where antagonists manage to construct a basis for consensus through a coordinative discourse (Schmidt, 2002). A coordinative discourse provides policy actors with a shared language, where they can agree on crucial elements of a reality description in a manner that enables them to agree on future policy actions. The Scandinavian preference for broadly composed (often tripartite) public commissions which author public reports on the state of crucial policy areas represents one way of facilitating such a coordinative discourse. Discursive effects are significant not only in political negotiation, but also at the micro level in organizations and public service units. The winds of New Public



Stability or Disintegration? 

 13

Management have swept over Scandinavia as they have over the rest of the modern world, albeit in the form of semi-manufacture, allowing for local adaptations (Røvik, 2007). This creates special problems of layering, where ideological and organizational considerations compete. Both practical solutions and ideological discourses may be characterized by compromises that threaten to become excessively complex and thereby counter-productive (Vabø, this volume). One important source of stability is the forms of flexibility which allow for the handling of crises with low transaction costs, while at the same time contributing to the reduction of unemployment. Danish “flexicurity” is one such system writ large. A relevant small-scale institution is that of temporary layoffs, where workers receive unemployment benefits during layoff periods even though the employment contract is sustained, allowing the worker to return to the same job when the economic situation of the firm has improved. This arrangement is an unusual compromise, again anchored in the tripartite system where the labour market organizations negotiate while the state bears the financial responsibility for the benefits (Olberg, this volume). If layoffs are unavoidable, they are handled quite differently in the Nordic countries. When crisis occurs, institutionalized rules of course matter, but at the same time they are interpreted and practiced with considerable discretion, based both on the distribution of resources between actors and the institutional context they operate within (Svalund, this volume).

1.4 Stability or Disintegration? How stable is the neo-corporatist system, after all? A system that to a large extent is built on legitimate order, compromise and trust among the parties also has significant sources of disequilibrium. When dialogue and internal security routines are central parts of the game, as in Scandinavia, it obviously becomes vulnerable to moral hazard or even outright criminal deviation from the existing rules. Nicolaisen and Trygstad (this volume) make the case that the inflow of migrants into the lowskilled industries, such as construction and cleaning, create opportunities for hiring people exempted from existing rules and agreements. Putting pressure on “decent” operators in the market, the emergence of such actors leads to the formation of stricter external control, thereby potentially undermining the dialogical regime. A possible long-term consequence is growing contagion from some specific sectors to wider parts of the economy. However, it is not only employers who may see it in their short-term interest to evade existing institutional regulation; another conceivable source of disintegration is a revolt from the middle class against large-scale collective bargaining and the ensuing compression of the wage structure (Moene & Wallerstein, 2003). In addition to the pressures from the labour markets, the viability of the Nordic/ Norwegian model is also threatened by uneven reform capacities. The pension

14 

 Introduction: Institutional Change in Neo-Corporatist Society

reform is an example of successful social bargaining, whereas the stalemate in the conflicts over sickness payment may be taken as a sign that corporatism sets up its own barriers to efficient reform. Maybe such barriers will remain until a drastic crisis opens a window of opportunity for change. Alternatively it may be that the institutional framework of the Nordic model is sufficiently robust to allow future adjustments without undermining its basic traits. With a view from outside, some perceptive observers point in this direction (Katzenstein, 2003; Thelen, 2014); others imagine a bleaker future (Crouch, 2004; Streeck, 2011). Making solid predictions hardly makes sense. However, the contributions to this book hopefully invite well informed reflection.

References 

 15

References Andersen, S.K., Dølvik, J.E & Ibsen, C.L. (2014). De nordiske aftalemodeller i åbne markeder, NorMod 2030, report 9. Oslo: Fafo Avdagic, S., Rhodes, M. & Visser, J. (2011). Social Pacts in Europe: Emergence, Evolution and Institutionalization. Oxford: Oxford University Press. Booth, M. (2014). The Almost Nearly Perfect People: The Truth About the Nordic Miracle. London: Jonathan Cape Boudon, R. (1991). Theories of Social Change: A Critical Appraisal. Cambridge: Polity Press. Brandal N., Bratberg, Ø & Thorsen, D.E. (2013). The Nordic Model of Social Democracy. Basingstoke: Palgrave Macmillan. Collier, D., & Collier, R. (1991).s Shaping the Political Arena. Critical Junctures, the Labour Movement and Regime Dynamics in Latin America. Princeton: Princeton University Press. Crouch, C. (2004). Post-Democracy. New York: Wiley. Dahrendorf, R. (1959). Class and Class Compromise in Industrial Society. Palo Alto: Stanford University Press. Demsetz, H. (1967). Toward a Theory of Property Rights. American Economic Review, 57, 347-359. Dølvik, J.E. (2007). The Nordic regimes of labour market governance: From crisis to success story? Fafo-paper 2007:07. Oslo: Fafo. Economist. (2013). The next supermodel. The Economist, Feb 2nd 2013. Esping-Andersen, G.(1985). Politics against markets: The social democratic road to power. Princeton: Princeton University Press. Esping-Andersen, G. (1990). Three Worlds of Welfare Capitalism. Princeton: Princeton University Press. Fasting, M. (2014). Har vi råd til fremtiden? Perspektivmeldingens utfordringer. Oslo: Civita. Hall, P & Taylor, R. (1996). Political Science and the Three New Institutionalisms. Political Studies, 4, 936-957 Hall, P. & Soskice, D. (2001). Varieties of Capitalism. Oxford: Oxford University Press. Häusermann, S., Picot, G. & Giering, D. (2012). Review Article: Rethinking Party Politics and the Welfare State – Recent Advances in the Literature, British Journal of Political Science, 43(1), 221-240. Katzenstein, P. (1985). Small States in World Markets. Ithaca: Cornell University Press. Katzenstein, P. (2003). Small States and Small States Revisited. New Political Economy, 8, 9-30. Korpi, W. (1983). The Democratic Class Struggle. London: Routledge & Kegan Paul. Mahoney, J. & Thelen, K. (2010). A Theory of Gradual Institutional Change. In J. Mahoney & K. Thelen, (Eds.), Explaining Institutional Change. Ambiguity, Agency, and Power. Cambridge: Cambridge University Press. Mann, M. (1986). The Social Sources of Power. Cambridge: Cambridge University Press. Martin, C.J. & Swank, D. (2012). The Construction of Business Interests. Cambridge: Cambridge University Press. Mitchell, D.J. (2007). What Can the United States Learn from the Nordic Model? Policy Analysis. November 5, 2007, No. 603. Moene, K. & Wallerstein M. (2003). Earnings Inequality and Welfare Spending. World Politics, 58, 485-516. Pierson, P. (2004). Politics in Time. History, Institutions, and Social Analysis. Princeton: Princeton University Press. Pierson, P. (2006). Public Policies as Institutions. In I. Shapiro, S.Skowronek & D. Galvin (Eds.), Rethinking Political Institutions. The Art of the State. New York: New York University Press.

16 

 References

Rokkan, S. (1967). Numerical Democracy and Corporate Pluralism. In R.A. Dahl (Ed.), Political Oppositions in Western Democracies. New Haven: Yale University Press. Røvik, K. A. (2007). Trender og translasjoner. Oslo: Universitetsforlaget. Schmidt, V. (2002). The Futures of European Capitalism. Oxford: Oxford University Press. Scott, R. (2008). Institutions and Organizations. Thousand Oaks: Sage. Sejersted, F. (2001). Capitalism and Democracy: A Comparison Between Norway and Sweden. In H. Byrkjeflot, S. Myklebust, C. Myrvang & F. Sejersted (Eds.), The Democratic Challenge to Capitalism. Bergen: Fagbokforlaget. Sejersted, F. (2011). The Age of Social Democracy: Norway and Sweden in the Twentieth Century. Princeton: Princeton University Press. Streeck, W. (2011). The Crisis of Democratic Capitalism. New Left Review, 71, 5-29. Streeck, W. & Thelen, K. (2005). Institutional Change in Advanced Political Economies. In W. Streeck & K. Thelen (Eds.), Beyond Continuity. Institutional Change in Advanced Political Economies. Oxford: Oxford University Press. Stråth, B. (2001). Nordic Capitalism and Democratization. In H. Byrkjeflot, S. Myklebust, C. Myrvang & F. Sejersted (Eds.), The Democratic Challenge to Capitalism. Bergen: Fagbokforlaget. Thelen. K. (1999). Historical Institutionalism in Comparative Politics. Annual Review of Political Science, 2, 369-404. Thelen, K. (2004). How Institutions Evolve: The Political Economy of Skills in Germany, Britain, Japan and the United States. Cambridge: Cambridge University Press. Thelen. K. (2012). Beyond Comparative Statics: Historical Institutional Approaches to Stability and Change in the Political Economy of Labor. In G. Morgan, J. Campbell, C. Crouch & O.K. Pedersen (Eds.), The Oxford  Handbook of Comparative Institutional Analysis (pp. 41-62). Oxford: Oxford University Press. Thelen, K. (2014). Varieties of Liberalization and the New Politics of Social Solidarity. Cambridge: Cambridge University Press.

Cathie Jo Martin

2 Negotiation and the Micro-Foundations of Institutional Change “It’s déja vu all over again.” Yogi Berra

The Nordic havens of social solidarity seem to defy the conventional “equality versus efficiency” trade-off (Okun, 1975). Close linkages between working life and the welfare state mean that expansive social spending is particularly targeted on investments in skills training; generous unemployment benefits do not lead to high unemployment; and equality coexists with productivity. Scandinavian citizens express significantly higher levels of trust in government, labor unions, and employers than their Anglo counterparts. Even through the global financial crisis, the Nordic countries have sustained low budget deficits, high employment, and a robust public sector. Perhaps even more surprisingly, Scandinavian countries periodically redesign welfare state policies and institutions in ways that challenge our understanding of the social democratic model and baffle students of institutional change. These sometimes profound institutional changes seem at odds with the policy legacies of the Nordic model, yet they often work to preserve social democratic outcomes of equality and solidarity. For example, the widespread adoption of active labor market policies in the 1990s greatly scaled back passive employment benefits and was viewed by many as an assault on the welfare state (Abrahamson, 1998). Yet the broader ambition of the proponents of activation was to reduce dualism, build skills and reincorporate marginal workers back into the core economy. This institutional reinvention of social assistance, seemingly so contrary to the historical practices of social democracy, was implemented in ways that resonated with the deep logics of social democracy (Cox, 2001; Martin, 2004; Martin & Swank, 2004; Martin & Thelen, 2007). In this volume, Mari Teigen’s tale of the legislation of gender quotas for corporate boards provides another case in point. Regulated mandates play an insignificant role in and are at odds with the policy legacies of the Norwegian industrial relations system, yet the new quotas fostering gender equality were supported by all parties apart from the populist far right party. Although the largest employers resisted the mandated solution, the political consensus prompted them to create the Female Future project, to identify women for future board positions. This chapter reflects on the institutional change processes that enable Norway and other Nordic countries simultaneously to redesign and preserve their social democratic solutions, and considers the implications of the Scandinavian slight-ofhand for broader theories of institutional change. In particular, the chapter sheds light © 2015 Cathie Jo Martin This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

18 

 Negotiation and the Micro-Foundations of Institutional Change

on three related mysteries of institutional change in Scandinava. First, how do these lands frequently produce reforms that offer substantial, pie-expanding, long-term benefits for a broad cross-section of their societies? Second, why do the Scandinavian countries appear to have greater capacities both for encompassing paradigm shifts and for intentional incremental institutional adjustments? Third, how do these countries sustain the essence of social democracy and a measure of continuity even within moments of significant institutional change? Familiar patterns of political engagement, relations among social actors, and outcome measures often persist even during dramatic institutional upheaval and radical departures from the policy status quo; therefore, one needs to explain the puzzling conundrum of institutional continuity within change (Pettigrew, Woodman, & Cameron, 2001, p. 697). The secret to the endurance of the social democratic model and institutional continuity within change lies with the rules of collective political engagement, or procedures for the negotiation of political reforms, that have an indelible impact both on types of policy pacts available to national decision makers and on institutional change processes. Diverse rules for collective political engagement influence participants’ perceptions of problems, their level of trust, and their willingness to place broad, long-term, often uncertain group interests above their own narrow concerns. Individuals fall prey to negotiation myopia, or psychological and strategic tendencies to make sub-optimal deals in collective bargains and to misunderstand their broad, long-term, collective interests (and this has been repeatedly demonstrated in laboratory experiments and elsewhere). For example, individuals often scrimmage over zero sum gains instead of seeking out and endorsing pie-expanding options with broader benefits; moreover, they generally value short-term benefits over longer-term ones. These myopia may bring negotiating parties to settle for suboptimal bargains and miss broader deals that bring benefits to both sides of the table (Mansbridge & Martin, 2013). The Nordic countries have developed institutions for political negotiation, such as industrial relations and party systems, which include rules for political engagement and work against negotiation myopia. For example, a strong role for non-partisan expertise over politicized information helps to foster a shared and broader conception of a problem and its solutions. Repeated interactions among negotiators across policy areas and time augment trust and punish participants for devious strategic behaviors. The institutional use of penalty defaults enhances the likelihood of action in a negotiated agreement. The development of policy in non-legislative, somewhat private settings helps to prevent public posturing and enables greater movement in negotiated positions. The procedural tools for overcoming myopia in political negotiations enable both more solidaristic public policy and the capacity for more far-reaching institutional changes. Robust, transformational changes and solidaristic policy reforms are more likely to occur when deviations from the status quo expand the pie rather than simply redistribute benefits, anticipate second order changes, gratify a wide



Models and Determinants of Institutional Change  

 19

spectrum of interests, rest on universal principals, and meet substantive (rather than simply political) goals. Of course, not all institutional changes are planned and we often find ourselves racing to adjust to crises or confronted with changes done for political reasons. Moreover, slow, institutional erosion processes may take us unawares until we discover that our expectations do not fit with reality (Streeck & Thelen, 2005). Yet institutions and rules of collective political engagement matter deeply to the persistence of even these unexpected changes, because political forums matter to subsequent implementation and refinement of the policy and institutional changes. Thus the forums and institutional structures that channel our responses to sudden crises or gradual decay play a role in institutional change and deserve our consideration.

2.1 Models and Determinants of Institutional Change Contemporary scholarship focuses on two types of institutional change processes: discontinuous punctuated equilibrium models of change and incremental transformations.1 Discontinuous punctuated equilibrium models view change as occurring at moments of major upheaval, dramatically transforming institutions, and establishing new paths for future political trajectories. Path-altering institutional changes are often associated with paradigm shifts, which entail deviations from the overarching goals of state intervention, rather than from levels of benefits or instruments of public policy (Hall, 1993, p. 280). In punctuated equilibrium models of change, strategic choices at critical junctures establish policy legacies for future action and define the new normal. Moments of economic and political upheaval permit a broad repertoire of response and early outcomes are unpredictable. Yet the resolution of political conflicts at critical junctures creates enduring path dependencies through lock-in effects and feedback processes. Sequencing matters enormously and the path dependencies clearly lay down a track for future policy incarnations (Capoccia & Ziblatt, 2010; North, 1990; Orren & Skowronek, 2004; Pierson & Skocpol, 2002; Baumgartner & Jones, 1993; Soss & Schram, 2007; Weir, 1993). Institutions created at critical junctures have subsequent reinforcing mechanisms including fixed costs, learning effects, coordination effects, and adaption expectations in technological exchange (North, 1990). Ideas are crucial to new trajectories in discontinuous institutional change, because strategic choices and legitimacy are influenced by cognitive paradigms and normative beliefs (Blyth, 2001; Campbell & Pedersen, 2014; DiMaggio & Powell, 1991; Hall, 1993; Schmidt, 2002).

1 Following Campbell, institutions are strategic equilibria, formal and informal rules, and norms as well as rules.

20 

 Negotiation and the Micro-Foundations of Institutional Change

Incremental institutional change processes unfold according to a very different logic, as these transpire over time and often underlie apparent institutional stability. Endogenous erosion processes such as drift, layering and conversion undermine the status quo. Although these processes sometimes lack intentional action, shifting coalitions of interests may be important to the conversion of institutions in incremental theories. The emergent coalitions of actors with new priorities seize control of institutions and change occurs through the erosion of institutional functions, conversion to new purposes, and layering of new goals onto prior ones (Hacker, 2002; Mahoney & Thelen, 2010; North, 1990; Streeck & Thelen, 2005; Thelen, 2004). Both punctuated equilibria and incremental theories of institutional change have made great strides in exploring the dynamics of policies and institutions in transition. In particular, these offered important corrections to early institutional analyses that accounted for enduring cross-unit differences but that often neglected agency, constructed preferences, and incremental change. Yet the new conventional wisdom in institutional analysis also leaves crucial gaps. Acute and incremental changes are often propelled by similar dynamics and may be sorted according to other qualities on a continuum (Baumgartner, 2013). For example, both acute and incremental changes vary as to whether reforms replace or add to extant institutions, are enduring or unstable over time, and are perceived as legitimate or illegitimate. First, the scope of change – encompassing versus additive – is crucial to whether new institutions replace the old and establish governing authority or simply duplicate the efforts of other organizations (Goldstein, 1988, p. 181; Orren & Skowronek, 2004, p. 86). Fragmented and incomplete additive institutional innovation offers more limited reform potential than encompassing change. For example, in the realm of incremental institutional changes, conversion is more likely to constitute replacement while layering simply adds to the status quo, fragments governing authority and compromises adjustment potential (Streeck & Thelen, 2005, p. 42). In this volume, Anniken Hagelund and Axel West Pedersen’s comparison of the pension and sick pay policy reforms illustrate the sharp contrasts between encompassing and additive reforms. Second, institutional reforms evidence varying levels of durability over time. In Orren and Skowronek’s discontinuous change model, transformational change happens at moments in which a country experiences a “durable shift in governing authority,” and true paradigm shifts require stable and robust alterations in our conceptualizations of a problem and its solution (Hall, 1993). The stability of institutional change is influenced, in part, by the degree to which the needs of future actors have been taken into account during the change process. Reformers have difficulty imposing short-term costs for long-term benefits, yet this capacity may influence the ultimate durability of the change (Mansbridge & Martin, 2013). Timebased sequencing of reforms also matter to the durability of an institutional change (North, 1990; Orren & Skowronek, 2004; Pierson, 2004; Pettigrew et al., 2001, p. 699). In incremental change processes, durability depends in part on the enduring strength of coalitions supporting the changes. Thus, Anne Lise Ellingsæter (in this volume)



Models and Determinants of Institutional Change  

 21

compares the durability of coalitional support for cash for childcare benefit schemes in Finland to the uncertain fate of the programs in Norway and Sweden. Third, institutional change may be perceived as more or less legitimate. The degree of perceived legitimacy depends, in part, on whether transformations seem motivated by “considerations of justice” and universal as opposed to particularistic principles, because changes rooted in universal considerations have greater perceived legitimacy (Stinchcombe, 1997, p. 5). Institutional reforms with positive impacts on a broader cross-section of citizens should gain greater legitimacy than those with a narrow impact. Change accepted as legitimate is more likely to be motivated by substantive and goal-oriented motives than political ones; and creative solutions that seem to rise above ideological divides may also enhance the perceived legitimacy of organizational change (Pettigrew et. al., 2001, pp. 699-700). In this volume, Mia Vabø delves into the legitimacy problems associated with New Public Management reforms in home care. Proponents drew from various discourses to advance and legitimize the reforms, and in the process, built top-down and bottom-up support for the measures. There is also an (albeit somewhat complicated) relationship between change processes and outcomes. Pettigrew et  al. (2001, p. 703) suggest that more encompassing replacement change – presumably that is also more durable and legitimate – is associated with better outcomes; for example, cross-national studies of organizational change in Europe suggest a “strong association between whole-system change and firm performance.” Table 1 presents the ways in which these diverse features characterize diverse types of institutional change. Table 2.1: Features of acute and incremental change* Encompassing/replacement

Additive

Durable/robust over time Paradigm shifts & legitimate Durable changes in Governing Authority Conversion Creation of VAT

Adjustment of policy tools Layering Medicare in the US

Durable but illegitimate Durable shifts in governing authority made by authoritarian regimes and reinforced with social control Nazi racial policies

Incremental changes made under cover of administrative decisions that lack social support but persist.

Unstable over time

New Policies and Institutions in Westminister system Tony Blair’s New Deal

* Institutional changes are in bold.

Drift Employment-based health benefits in US

22 

 Negotiation and the Micro-Foundations of Institutional Change

The dominant institutional change theories account for variations over time, yet these do not address sources of continuity within change, such as the perseverance of social solidarity within the social democratic model. For example, in this volume, Nicolaisen and Trygstad describe a surprising policy change in response to the stresses on solidarity posed by new, low-end service workers. Political actors have fought the growing worsening economic plight of these workers with an institutional innovation, hard regulations to combat illegal practices, which seem completely at odds with the historical delegation of control over industrial policies to the social partners. The authors reflect on whether this regulatory solution constitutes a failure of the Nordic model, but also recognize that this adaptation enables the continuing social democratic commitment to good working conditions. Even while the tools of intervention seem inconsistent with the Nordic model, the spirit of social democracy persists as a source of continuity within institutional change. The dominant models of institutional change also pay scant attention to the micro-foundations of institutional change, or the factors bringing reformers to hold preferences for change. In part, this reflects a tension between the historical institutionalist emphasis on the institutional influences of group preferences and the rational choice theorist view of individual preference as grounded in rationallyappraised economic circumstances. Scholars offer “agency” to account for variations in new institutional paths: agency guides idiosyncratic decisions at critical junctures and influences the composition of coalitions that propel incremental change. Yet agency has a transitory quality – a black box with little predictive power – and new paths and coalitions are seldom entirely random. In the following sections, I explore how processes for collective political engagement may exert influence on both discontinuous and incremental institutional change. Variations in the rules of political engagement shape the manner in which policy legacies are reinterpreted at critical junctures. These variations in processes of engagement delimit the range of political coalitions that might contribute to incremental institutional change. The institutions and rules of collective political engagement also have an impact on the implementation of and retrospective adjustments to new institutional creations, in their impacts on the durability and legitimacy of institutional reforms.

2.2 The Micro-Foundational Underpinnings of Institutional Change To grasp the Nordic successes, one needs to move beyond the conventional wisdom and explore the micro-foundations of institutional change. This entails looking at how rules of engagement influence the negotiation myopia of participants in the reform process and, under some circumstances, inspire these participants to engage in deliberative negotiation. This section defines “deliberative negotiation” and “zero-



The Micro-Foundational Underpinnings of Institutional Change 

 23

sum bargaining,” delves into the psychological and strategic myopia that hinder deliberative negotiation, and considers the procedural rules that aid in overcoming myopia. Finally, we consider how lessons derived from dynamics of negotiations inform our thinking about institutional change in general and the Norwegian case in particular.

2.2.1 Prototypes of Negotiation The negotiation literature differentiates between types of negotiation and considers the procedural arrangements that foster these diverse types. In a moment, we will consider how this distinction has a parallel in types of national political deals. At one extreme, participants engage in antagonistic bargaining, view outcomes as zerosum, and worry more about claiming a great share of the distribution of benefits than about expanding the total pie. At the other extreme, participants enter into inclusive or deliberative negotiations, in which the negotiating partners engage in consensual discussions in efforts to obtain collectively beneficial outcomes that create value or expand the proverbial pie (Dür & Mateo, 2010, p. 682).2 In “distributive bargaining” situations, agreements meet minimal process requirements of efficiency and utilityenhancement; however, outcomes are sub-optimal (as they do not expand the pie), entail distributive, zero-sum exchanges and particularistic pay-offs, exclude the interests of those not represented at the table, and neglect long-term consequences. In “inclusive negotiations,” agreements meet the minimum process requirements, but also produce optimal outcomes that make a majority of people better off. Participants act toward collective instead of individualistic interests, think about long-term impacts on future generations, and focus on substantive rather than political goals. While difficult to measure, these concepts suggest a hierarchy of intervention that is intuitively known to us. A begrudging zero-sum “horse-trading” is qualitatively different from those moments when thinking outside the black box enables us to arrive at a creative solution that meets with collective needs and interests broadly defined (Bellamy, Kornprobst & Reh, 2012). Dag Olberg in this volume provides an example of a win-win deal, negotiated by Norwegian social partners to protect workers in the case of economic crisis. Both sides wished to develop a tactic that would spare workers the humiliation of being laid off and reduce economic pressures on firms facing reduced demand for their products. Their win-win solution was the “temporary layoff,” in which the state would pay partial support for the workers but the workers could cling to their employment status

2 Descriptions of “bargaining” and “inclusive negotiation” include: “value-claiming versus valuecreating,” “distributive versus integrative bargaining,” “bargaining versus problem-solving,” “strategic action versus communicative action,” and “hard versus soft bargaining.”

24 

 Negotiation and the Micro-Foundations of Institutional Change

and security, because they remained employees of the firms. Table 2 presents these models of collective deal-making. Table 2.2: Distributive bargaining versus inclusive bargaining Distributive Bargaining

Inclusive Negotiation

Outcomes Measures

meet minimum process criteria, e.g not devious, all better off, but suboptimal

meet process criteria, but also optimal outcomes, even for those not at table

Object

distributive wins value claiming, political goals

pie-expanding solutions value creating, substantive goals

Scope of interests individual interests

collective interests

Time-line

long-term gains

short-term gains

2.2.2 Negotiation Myopia The essential question, of course, is why individuals – who stand to win much more from inclusive negotiation – are more often likely to engage in redistributive bargaining. In part, this is because the human brain falls prey to negotiation myopia, or cognitive, psychological, and strategic impulses that prevent individuals from reaching agreements that could conceivably leave all better off. First, we experience myopia in our perspective of ourselves vis-à-vis others. Due to a self-regarding bias, we interpret events from our own vantage point, but an alternative other-regarding perspective would help us to grasp courses of action that could improve the lot of the whole, often at minor discomfort to ourselves. With information asymmetry, because we do not have full knowledge of others’ motives, we may falsely assume that our preferences are incompatible with those of our opponents (Pronin, Lin & Ross, 2002; Bazerman et. al., 2002). Second, we experience myopia in terms of the scope of goals. Our concerns about the distributive of benefits make us blind to the benefits of joint action that could enlarge the pool of resources and create new value. Loss aversion may enhance the tendency toward distributive conflict, in making us risk adverse, even when the potential pay-offs are much greater than the possible losses. For example, unions that refuse to consider productivity-enhancing technological improvements may save the costs of retraining for some members but may ultimately make the company lose market share. Individuals also may fall prey to reactive devaluation, in which they tend to discount advantageous deals that are offered too readily (Mnookin & Ross, 1995, p. 17).



The Micro-Foundational Underpinnings of Institutional Change 

 25

Third, we may suffer from time myopia that diminish our capacities to grasp the long-term perspective and consider second and third order effects. Many public policies – for example, pensions – require short-term costs or investments in order to achieve longer-term benefits and this accentuates our tendency to overvalue the present, particularly when concentrated interests must bear a greater share of the costs but will not be able to claim a proportionate share of the future benefits (Jacobs, 2011, p. 52). Our tendency to embrace the short-term is complicated by our difficulty understanding second or third order effects. We often derive predictions about the future from easily accessible cases, imperfect data, and narrowly circumscribed considerations.

2.2.3 Rules to Overcome Negotiation Myopia Certain procedural arrangements to thwart negotiation myopia may be introduced into private negotiations to achieve more optimal outcomes, and corresponding rules for collective engagement in national policy processes may likewise increase the likelihood of developing broadly-inclusive, enduring, pieexpanding political deals. These arrangements influence conceptualizations of problems and solutions, attitudes toward cooperation, and incentives for action. These rules may also be adopted within a subnational policy area or a specific political debate. The first rule that may foster inclusive negotiation is an incorporation of nonpartisan expertise into the deliberative process. Nonpartisan experts help participants to overcome self-serving biases in the perception of facts, foster a shared understanding of policy problems in more neutral terms, build shared conceptions of justice, diminish ideological left-right cleavages, and enable creative cognitive leaps. Countries develop characteristic “knowledge regimes” that structure policy learning, and these exhibit varying degrees of reliance on nonpartisan expertise, which contributes to the diverse modes of discourse (Blyth, 2001; Campbell & Pedersen, 2014; Schmidt, 2002). Crucial forums for nonpartisan expertise may also contribute to breakthrough agreements in specific policy areas, as was the case when an important governmental commission established support for a new child care system in the Netherlands (Morgan, 2006). Torgeir Nyen and Anna Hagen Tønder (in this volume) describe the crucial meeting of the minds among professional educators and the social partners in a major Norwegian vocational training reform in 1994. Although school and apprenticeships were only weakly connected before the reform, the national educational authorities issued important research on the problematic transition of youth to work and skills deficits among Norwegian workers. Exposed to the new model of vocational training within their encompassing associations, employers and workers fully supported the wholesale creation of a new dual VET system. In contrast, Cathrine Holst (also in this

26 

 Negotiation and the Micro-Foundations of Institutional Change

volume) finds sharp distinctions between academics’ and employers’ views on the equal pay controversy. The discrepancies among these views reflect the institutional context. Second, a negotiation context that includes repeated interactions help to foster other-regarding perspectives, win-win solutions, and long-term thinking. Classic problems in prisoner’s dilemma games are distrust and incentives for short-term, self-interested, sub-optimal behavior; however, these are reduced by iterative interactions, in which current actions have future implications (Axelrod, 1984; Hardin, 1968; North, 1990; Olson, 1965). According to Eivind Falkum (in this volume), repeated interactions helped to build trust and shared understanding after the establishment of the Norwegian Basic Agreement of 1935 between the employer and labor confederations. The Basic Agreement was born out of industrial conflict; however, through repeated engagements the social partners came to recognize the political and economic advantages of cooperation and converged on many ideological and normative assumptions. Third, the use of penalty defaults establishes a mandate for action and greatly expands the likelihood that an agreement will be reached. Procedural rules and norms that include the provision by external agents of penalty defaults, exclusion from the table, inducements, or other action-forcing rules (external conditionalities) may help to produce negotiation processes and outcomes.3 These defaults may be structured to use costs distributed over the medium term for long-run social benefits. Uncertainty about future economic and social conditions makes it more difficult for participants to assume future burdens and this, more than short-term thinking, diminishes support for longer-term solutions (Jacobs, 2011). Penalty defaults may be necessary to force negotiating partners to adhere to outcomes, even when repeated interactions have built trust and long-term relationships. For example, Norwegian managers adhere to dismissal regulations in quite varied fashion, and this level of discretion reflects the power of local governments to enforce compliance at the implementation stage and the strategic power and attitudes of corresponding trade associations (Jørgen Svalund in this volume).

3 These penalty defaults may simply produce a the most elemental minimally utility-enhancing type of negotiation, as when a fear of stalemate motivates actors simply to impose long-term costs on shortterm benefits, which may benefit those at the table but harm groups not represented in the negotiations. Yet these defaults might also be structured to motivate processes and decisions that are more other-regarding, pie-expanding and long term.



The Micro-Foundational Underpinnings of Institutional Change 

 27

2.2.4 Rules for Collective Engagement and the Institutions of Consensual Democracies The rules for collective political engagement are embedded in the institutions of consensual democracies, and these account for the heightened capacities for cooperation within these institutional settings. In particular, macro-corporatist industrial relations systems and proportional party systems make use of the procedural arrangements that foster inclusive negotiations and these account for the greater prevalence of inclusive agreements in coordinated market economies. Proportional party systems, compared to majoritarian ones, are typically (but not always) programmatic rather than patronage parties and organize constituents around substantive rather than purely political ends. These programmatic parties typically nurture technical expertise in their units for policy development, in contrast to patronage parties in countries such as the United States, where interest groups craft most legislation and constituents are linked to their parties through material benefits rather than with ideas. Consensual democracies require greater cross-party negotiation and are more likely to rely on neutral commissions to assess policy issues with broad cross-party and cross-class representation. Representatives of proportional parties are more likely than those of majoritarian parties to engage in repeated interactions with one another, because in proportional, multiparty systems (with rare single-party majority rule), parties must cooperate in order to form a governing coalition and to enact legislation. The need to form coalition governments also constitutes a penalty default for participants to arrive at a compromise. Finally, because proportional parties usually have more complete coverage of their constituency groups, they tend to compete less for marginal voters and may more easily make credible commitments that inspire trust in longer-term institutional changes (Cusack, Iversen & Soskice, 2007; Kitschelt, 1993). The crucial role for technical experts in enabling broad social pacts is illustrated by the use of royal commissions in Sweden, such as the expert task force on climate change in the 1970s, which set the stage for early clean air legislation. Although the Swedish bill was less extensive than parallel legislation in the United States, its impact was far more substantial because the legitimacy established through the expert investigation made for easy implementation and extensive compliance (Lundqvist, 1980). Macro-corporatist industrial relations systems also make extensive use of the arrangements facilitating inclusive negotiation in their structural design. Countries with macro-corporatist industrial relations systems include a more formal role for technical expertise in the tripartite commissions that bring business, labor and the state together to consider policy problems, as these forums nurture shared understandings of problems and solutions. Employers have a range of possible interests, multiple objectives, and many intermediate goals; highly-organized business associations educate employers and bring managers into contact with policy experts from government and organized labor. Repeated interactions are an

28 

 Negotiation and the Micro-Foundations of Institutional Change

important feature of both collective bargaining processes and participation on tripartite commissions found in macro-corporatist industrial relations systems. In macro-corporatist industrial relations systems, a penalty default appears in the state’s threat to intervene if social partners do not reach agreement. The social partners lose control of their policy making prerogatives and power reverts back to the state in the absence of action (Martin & Swank, 2012).

2.2.5 Rules for Collective Political Engagement and Institutional Change Finally, institutional arrangements enabling inclusive policy negotiations also have bearing on modes of the discontinuous and incremental institutional change processes presented in Table 1. First, change may be encompassing in replacing old institutions with new ones or additive, a distinction that applies to both paradigm shifts in punctuated equilibrium models and conversion in incremental models of change. An institutional role for the influence of neutral technical expertise makes a strong contribution to countries’ capacities to replace older institutions with new ones. Specially-appointed bodies of technical experts to analyze problems and offer new policy solutions often have the intellectual capacities to grasp broad social and economic problems and the knowledge of emergent paradigms for coping with such problems. Moreover, their formal mandate facilitates an acceptance of new paradigms, which makes it easier for legislators to justify redesigning policies wholesale rather than simply adding new programs onto existing ones. Forums for iterative interactions among representatives of diverse social interests may help to popularize the new paradigms and to secure broader support for their political implementation, and this is particularly true with the technical experts representing the social groups who have played a role in the commissions responsible for developing the new paradigm for action. Penalty defaults in the form of a widely-perceived need or formal mandates to take action work against the default options of doing nothing and stalemate. For example, the solidaristic active labor market reforms in Denmark constituted a new paradigm for anti-poverty programs; whereas the German version of ALMP simply added onto the existing programs and constituted only a minor shift from the status quo (Martin & Swank, 2012; Martin & Thelen, 2007). The Danes were able to engineer these profound changes in social provision with arrangements that foster inclusive negotiation. The intellectual basis for the new programs was developed in two expert commissions (such as the Zeuthen Commission for the social insurance interventions), which studied and linked together the issues of long-term structural unemployment and a growing shortage of skilled labor. This reliance on technical expertise helped to introduce new paradigm of social provision into dialogue and to link this paradigm to the post-industrial growth model: an encompassing labor market could be created to include all workers, programs for the development of



The Micro-Foundational Underpinnings of Institutional Change 

 29

human capital should tailor interventions to individuals’ capacities rather than their incapacities, and the state should train to provide the supply of workers for specific demand needs of employers. The Danes used iterative interactions to foster acceptance of these new ideas across a broad cross-section of society, by formally involving the social partners in the expert commissions and by cultivating a common understanding of the problems. They also used penalty defaults to push through the reforms: The Danish Social Ministry told the major employers’ association and peak union that if the social partners did not develop a specific plan for implementing the new model, the state would both cultivate relations with individual firms and infringe upon the traditional control over industrial relations policy by the social partners (Martin, 2004; Martin & Swank, 2012). In sharp contrast, German interventions constituted a much less stark formal institutional change than the Danish system, but the incapacities of policy-makers to engineer transformational change permitted significant institutional drift despite formal perceptions of institutional stability. In part, the differences lay with the institutional capacities for iterative interactions in the two countries. Institutions for Danish macro-corporatism build in iterative interactions among a broad crosssection of social partners, which enabled acute institutional replacement rather than incremental institutional drift. Sectoral coordination in Germany made crossclass coalitions at the industry level more likely than broad social pacts across the economy. This separated labor market insiders from outsiders, worked against broad institutional changes, and permitted dualism and institutional drift, as marginal workers increasingly lost the benefits of social protections enjoyed by the industrial elite. The monumental Danish reforms preserved the spirit of social democracy, even as they deviated from the status quo; whereas the incremental German adjustments chipped away at prior solidaristic arrangements, even as these preserved the appearance of institutional stability (Martin, 2004; Martin & Swank, 2012; Martin & Thelen, 2007; Thelen, 2014). Second, institutional change may be characterized as stable and robust or unstable over time, and here again the procedural arrangements facilitating inclusive negotiations also contribute to more robust institutional changes. A strong role for neutral technical expertise in policymaking processes enhances the capacities of participants to think through the ramifications of various courses of action and to set a priority on substantive goals over political rent-seeking benefits. This expansion of the level of dialogue makes participants more likely to take into consideration second and third-order effects of their actions and to focus on longer-term concerns. The existence of penalty defaults forces action and works against the likelihood of gridlock as the default action and institutional drift, in which the absence of intervention allows current institutional arrangements to become increasingly less capable of serving their intended role. Procedures for iterative interactions among representatives of a wide range of societal interests also contribute to the stability of institutional innovations over time, because these forums should bring people to

30 

 Negotiation and the Micro-Foundations of Institutional Change

trust one another and to rest assured that if they express a willingness to pay higher short-term costs in exchange for longer-term benefits, others will also contribute to the effort. Contemporary pressures in pension policies constitute a classic example of the difficulties involved in creating long-term, robust institutional change. The pay-asyou-go pension plans were initially set up under conditions of rapid population and economic growth; therefore, using the contributions of current workers to support current retirees was an easy solution to elderly poverty. But with changing demographic structures and declining worker to recipient ratios, experts agree that a capital investment approach is much more feasible, entailing workers to contribute to their own future retirement needs. Changing the system, however, requires difficultto-make short-term investments, because current workers must simultaneously continue to support current retirees and begin investing in their own individual funds. Although negotiators worry about the long-term ramifications, short-term solutions are often chosen due to uncertainty, limits to their cognitive perceptions, and fears of the unknown. In the Canadian case, however, policy makers used nonpartisan expertise to build a shared understanding of the problem, and relied on guarantees of cost sharing and penalty defaults to push for enduring institutional change (Jacobs, 2011, p. 52). A third characteristic of institutional change concerns whether changes are perceived as legitimate or illegitimate, and once again, a reliance on technical expertise, iterative interactions and penalty defaults all help to expand legitimacy. Perceived legitimacy depends on three related factors: a) the perceptions that the institutional changes are universal in their application rather than particularistic, b) the focus on substantive rather than political goals and c) the scope of the group whose interests are benefitted by the change, i.e. a broader group translates into a higher perception of legitimacy. A formal role for neutral technical expertise increases the likelihood that the intended impacts of institutional change will be substantive, encompassing, and universally applied, rather than guided by political quid-pro-quo exchanges. Involving a wide cross-section of interests in repeated discussions of the problem should help to expand the legitimacy of the institutional change process, and penalty defaults should, again, depress the likelihood of non-action. In this volume, Fredrik Engelstad recognizes the necessity of securing legitimacy in his discussion of far reaching reforms to corporate boards, which constituted a substantial intrusion of state policy into private business control. Engelstad suggests that the state walks a fine line between protecting property rights and securing democratization in policies to secure employee representation and to improve the gender balance on corporate boards. Policy makers secured this legitimacy, in part, by working through the macrocorporatist industrial relations institutions to build significant support among the social partners long before the point of legislation. Of course, technical expertise, iterative interactions, and penalty defaults do not always produce more encompassing, robust and legitimate changes. These rules of

Conclusion 

 31

engagement in isolation may worsen capacities to arrive at inclusive negotiations and broader institutional change. For example, iterative interactions in the form of high levels of societal coordination may be used to achieve authoritarian ends, a point Berman (1998) vividly makes in her analysis of features of German civil society. Moreover, coordination among narrowly circumscribed interests, absent a strong role for technical expertise and without penalty defaults to arrive at broadly applicable solutions, may allow for insider groups to derail efforts at broader reforms. Iterative interactions can also exacerbate ill feelings among negotiators and although coordination may produce more efficient and egalitarian outcomes in the core industrialized world, this may lead to rent-seeking behavior in the developing world, particularly when this exists among vertical linkages (Granovetter, 1985; RodriguezPose & Storper, 2006). Embedded stability in one domain may prevent broader change; thus, North (1990, p. 83) suggests that institutions have a “complex set of constraints that include formal rules nested in a hierarchy, where each level is more costly to change than the previous one.” Of course, a tension may compromise the joint achievement of process and outcome goals. The expectations of broad consensus require the consent of a larger group of participants than simple bargaining and there are times when more ambitious outcomes may require the political capacity to limit the objections of the range of negotiating partners. Yet when iterative interaction is combined with a formal role for technical expertise and penalty defaults to force negotiation, or institutional reforms develop that establish the capacities for inclusive negotiations, the dark side of coordination surfaces less easily.

2.3 Conclusion This chapter asks why liberal and coordinated countries often demonstrate an essential familiarity in their national responses to problems, even at moments of tremendous upheaval and radical change. The central argument of this paper is that the source of continuity within change has to do with processes of collective engagement, or the signature way that people come together to solve political problems. These are particularly important in their impacts on how individuals conceptualize their interests and negotiate outcomes, because diverse institutions and rules of engagement have different effects on the myopia that keep individuals from making pie-expanding bargains. Certain cognitive and strategic myopia hold individuals back from pie-expanding negotiations: these include the tendencies toward self-regarding assumptions, engagement in distributive versus pie-expanding bargaining, and preferences for short-term versus long-term solutions. Some institutions and rules of engagement overcome these myopia, influence policy and affect institutional change mechanisms because they have distinctive impacts on individual preferences and capacities to overcome negotiation myopia:

32 

 Negotiation and the Micro-Foundations of Institutional Change

Nations evolve characteristic ways of solving social and economic problems, which shape how actors negotiate new policies and engage in institutional innovations. While these processes are themselves a work-in-progress, a characteristic manner of problem-solving – informed by formal institutions for involving actors in political decisions as well as meso level rules and norms – seems to endure, even as it evolves, across epochs. These meta institutions or the ways that people come together to renegotiate new policy and institutional solutions shape the possibilities for diverse types of reforms. The most robust institutional changes are likely to occur when deviations from the status quo expand the pie rather than simply redistribute benefits, anticipate second and third order changes, gratify a wide spectrum of interests, rest on universal principles, and meet substantive (rather than simply political) goals. These micro foundational impacts reveal the logic of why institutions for consensual democracy seem better able to engage in disjunctive institutional changes that are encompassing (replacement versus additive), robust over time and perceived as broadly legitimate; and industrial relations systems and party systems – key forums for collective political engagement – seem particularly relevant. Institutional alterations are more likely to become disjunctive, encompassing, robust over time, and legitimate when these are undertaken through inclusive negotiations.

References 

 33

References Abrahamson. P. (1998). Efter velfærdstaten: Ret og pligt til aktivering. Nordisk Sosialt Arbeid, 18(3), 133-42. Axelrod, R. (1984). The Evolution of Cooperation. New York: Basic Books. Baumgartner, F. R. (2013). Ideas and policy change. Governance: An International Journal of Policy, Administrations, and Institutions, 26(2), 239-258. Baumgartner, F. R., & Jones, B. (1993). Agendas and Instability in American Politics. Chicago: University of Chicago. Bazerman, M. H., Curhan, J. R., & Moore, D. A. (2002). The Death and Rebirth of the Social Psychology of Negotiations. In G. Fletcher & M. Clarks (Eds.), Blackwell Handbook of Social Psychology: Interpersonal Processes. Malden: Blackwell Publishers. Bellamy, R., Kornprobst, M., & Reh, C. (2012). Introduction: Meeting in the middle. Government and Opposition, 47(3), 275-295. Berman, S. (1998). The social democratic moment: Ideas and politics in the making of interwar Europe. Cambridge: Harvard University Press. Blyth, M. (2001). The transformation of the Swedish model: Economic ideas, distributional conflict, and institutional change. World Politics, 54(1), 1-26. Campbell, J.L. & Pedersen, O.K. (2014). The National Origins of Policy Ideas: Knowledge Regimes in the United States, France, Germany, and Denmark. Princeton: Princeton University Press. Capoccia, G., & Ziblatt, D. (2010). The historical turn in democratization studies: A new research agenda for Europe and beyond. Comparative Political Studies, 43(8-9), 931-968. Cox, R. H. (2001). The social construction of an imperative: Why welfare reform happened in Denmark and the Netherlands but not in Germany. World Politics, 53(3), 463-498. Cusack, T. R., Iversen, T., & Soskice, D. (2007). Economic interests and the origins of electoral systems. American Political Science Review, 101(3), 373-391. DiMaggio, P. J., & Powell, W. W. (1991). Introduction. In P. J. DiMaggio & W. W. Powell (Eds.), The new institutionalism in organizational analysis. Chicago: The University of Chicago Press. Dür, A., & Mateo, G. (2010). Choosing a bargaining strategy in EU negotiations: Power, preferences, and culture. Journal of European Public Policy, 17(5), 680-693. Goldstein, J. (1988). Ideas, institutions, and American trade policy. International Organization, 42(1), 179-217. Granovetter, M. (1985). Economic action and social structure: The problem of embeddedness. American Journal of Sociology, 91(3), 481-510. Hacker, J. S. (2002). The divided welfare state: The battle over public and private social benefits in the United States. New York: Cambridge University Press. Hall, P. A. (1993). Policy paradigms, social learning, and the state: The case of economic policymaking in Britain. Comparative Politics, 25(3), 275-296. Hardin, G. (1968). The Tragedy of the Commons. Science, 162(3859), 1243–1248. Jacobs, A. M. (2011). Governing for the long term: Democracy and the politics of investment. New York: Cambridge University Press. Kitschelt, H. (1993). Comparative historical research and rational choice theory: The case of transitions to democracy. Theory and Society, 22(3), 413-427. Lundqvist, L.J. (1980). The hare and the tortoise: Clean air policies in the United States and Sweden. Ann Arbor: University of Michigan Press. Mahoney, J. & Thelen, K. (2010). A Theory of Gradual Institutional Change. In J. Mahoney & K. Thelen (Eds.), Explaining Institutional Change. Ambiguity, Agency, and Power. Cambridge: Cambridge University Press.

34 

 References

Mansbridge, J. & Martin, C.J. (Eds.). (2013). Negotiating Agreement in Politics. Report of the Task Force on Negotiation Agreement in Politics. Washington, D. C.: American Political Science Association. Martin, C. J. (2004). Reinventing welfare regimes employers and the implementation of active social policy. World Politics, 57(1), 39-69. Martin, C. J., & Swank, D. (2004). Does the organization of capital matter? Employers and active labor market policy at the national and firm levels. American Political Science Review, 98(4), 593-611. Martin, C. J., & Swank, D. (2012). The Political Construction of Business Interests: Coordination, Growth, and Equality. New York: Cambridge University Press. Martin, C. J., & Thelen, K. (2007). The state and coordinated capitalism: Contributions of the public sector to social solidarity in postindustrial societies. World Politics, 60(01), 1-36. Mnookin, R.H., & Ross, L. (1995) Introduction. In K. Arrow, R. Mnookin, L. Ross, A. Tversky, & R. Wilson (Eds.), Barriers to conflict resolution. New York: Norton. Morgan, K. (2006). Working Mothers and the Welfare State. Palo Alto: Stanford University Press. North, D. C. (1990). Institutions, institutional change and economic performance. New York: Cambridge University Press. Okun, A.M. (1975). Equality and Efficiency. The Big Tradeoff. Washington, D. C.: The Brookings Institution. Olson, M. (1965). The Logic of Collective Action. Cambridge: Harvard University Press. Orren, K., & Skowronek, S. (2004). The search for American political development. New York: Cambridge University Press. Pettigrew, A. M., Woodman, R. W., & Cameron, K. S. (2001). Studying organizational change and development: Challenges for future research. Academy of Management Journal, 44(4), 697-713. Pierson, P. (2004). Politics in time. History, institutions, and social analysis. Princeton: Princeton University Press. Pierson, P., & Skocpol, T. (2002). Historical institutionalism in contemporary political science. In I. Katznelson & H. Milner (Eds.), Political science: State of the the disipline (pp. 693-721). New York: WW Norton. Pronin, E., Lin, D. Y. & Ross, L. (2002). The bias blind spot: Perceptions of bias in self versus others. Personality and Social Psychology Bulletin, 28(3), 369-381. Rodríguez‐Pose, A., & Storper, M. (2006). Better rules or stronger communities? On the social foundations of institutional change and its economic effects. Economic geography, 82(1), 1-25. Schmidt, V. A. (2002). The futures of European capitalism. Oxford: Oxford University Press. Soss, J., & Schram, S. F. (2007). A public transformed? Welfare reform as policy feedback. American Political Science Review, 101(1), 111-127. Stinchcombe, A. L. (1997). On the virtues of the old institutionalism. Annual Review of Sociology, 23(1), 1-18. Streeck, W., & Thelen, K. (2005). Introduction: Institutional change in advanced political economies. In W. Streeck & K. Thelen (Eds.), Beyond continuity: Institutional change in advanced political economies. Oxford: Oxford University Press. Thelen, K. (2004). How institutions evolve. New York: Cambridge University Press. Thelen, K. (2014). Varieties of Liberalization and the New Politics of Social Solidarity. Cambridge: Cambridge University Press. Weir, M. (1993). Politics and jobs: The boundaries of employment policy in the United States. Princeton: Princeton University Press.

PART I: Working Life Institutionalization – Negotiating Large Scale Social Change

Fredrik Engelstad

3 Property Rights, Governance, and Power Balances When focusing on the specific forms of capitalism and working life in Scandinavia, two apparent paradoxes come to the fore. One springs out of the notion of “egalitarian capitalism”, launched by Kathleen Thelen (2014) to describe one of the trajectories in the European coordinated market economies. The notion of egalitarian capitalism may sound like an oxymoron; how can employment relations be egalitarian? To some extent they can; if not in an absolute sense, at least relative to other societies. Egalitarianism in Thelen’s version rests on a combination of solidaristic redistribution and a high degree of unionization and employer coordination. But coordination is not a sufficient precondition for egalitarianism; she points out that it also has “to do with variation over time in the scope of employer coordination and the purposes to which these coordinating capacities were put” (2014, p. 10). This in turn is dependent on “the political coalitions on which … institutions rest” (ibid.). This conception may be contrasted with the notion of “democratic capitalism” coined by historian Francis Sejersted (1991) in his analyses of the specific development of capitalism in Norway in the 20th century. Particular for Norwegian exceptionalism – Sonderweg in his term – is the industrial development via small-scale manufacturing, often located in relatively small local communities with moderate social distance between workers and employers. Is it possible to reconcile these contrasts – large-scale coordination with smallscale local communities? This may be possible along two lines. The first relates to Thelen’s linking of solidarity to the scope and purpose of coordination. Norms based in closely knit communities is one salient factor influencing both scope and purpose in the direction of solidarity. Another line is connected to the position of property rights in Scandinavia. Here a second paradox appears. According to international rankings, the Scandinavian countries are rated best in the world concerning security of property rights; Finland comes out on top, with Norway and Sweden in second place along with New Zealand. In comparison, the United States is ranked number 17 of 97 countries (International Property Rights Index, 2014). A paradox occurs when this is combined with the observation that property rights in business life are more flexible in Scandinavia than in most, if not all, capitalist countries. In her analysis of egalitarian capitalism, Thelen (2014) focuses on the institutional specificities of industrial relations, vocational training and labour market policies in Scandinavia by drawing comparisons to other European countries. Central to her conception is the power distribution and political choices for the maintenance of institutions. Institutional stability is not upheld automatically, but is maintained by a coalition of actors. Well organized employers, a strong trade union movement and a supportive state together form a stable, but shifting, tripartite relationship. Even if power matters, the actors must be oriented towards creating new solutions to © 2015 Fredrik Engelstad This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Mechanisms of Institutional Change 

 37

problems emerging as a consequence of continuous social change (Thelen, 2014, pp. 293-306). The present chapter aims to extend this perspective by exploring the combination of solidarity, organizational coordination and property rights. The empirical focus is on Norway, which is mentioned only in passing in Thelen’s book (2014). An underlying assumption is that property rights constitute a core element in the interpretation of institutional changes in modern economies. Thereby a question of the adequacy of the power-resource theory (Korpi, 1989), which is Thelen’s main reference to conceptions of power, is also raised. The Basic Agreement, signed in 1935, serves as point of departure (a similar agreement was signed in Sweden in 1938 [Lundh, 2010], and in Denmark it was established by 1899 [Due & Madsen, 2010; Martin & Swank, 2012]). It serves as background for the discussion of reforms adopted – or rejected – much later, from the 1970s and on. Common to these reforms is that they generated changes in the organizational patterns of capitalist enterprises; moreover, they challenged, and to some extent modified, conceptions of property rights. The empirical cases discussed are the introduction of the Working Environment Act of 1977, the revision of the Limited Companies Act of 1972 allowing employee representatives to demand a seat on the board of directors of joint-stock companies, and the changes in the Public Limited Companies Act of 2003 introducing gender quotas to the board of directors.

3.1 Mechanisms of Institutional Change Theories of institutional change may crudely be divided into two opposed types. The theory of punctuated equilibrium assumes that social order normally remains stable – depending on the paths chosen in the past – until an exogenous shock appears, and a new policy emerges. The approach of Thelen and colleagues (Thelen 1999, 2004; Streeck & Thelen, 2005; Mahoney & Thelen, 2010) represents an alternative perspective resting on the assumption that institutional change is going on continuously, due to the aggregate effects of a multitude of actions. Over time they result in changes in social patterns and even affect the design of institutions. Now, Norwegian political history does not quite fit any of these conceptions; it is rather the history of continuous reform endeavour, where specific policy developments and continuous change are equally important. Hence, the present chapter situates itself between the two opposed theoretical approaches, focusing on long-term institutional changes enacted by agreements and legislation just as much as by aggregates of action. However, this modification of perspective does not make Thelen’s typology of processes of institutional change less pertinent. Four types, or maybe rather four bundles, of mechanisms generating gradual institutional change are distinguished: Displacement, Layering, Drift and Conversion. The two former relate to the introduction of new rules to the institution, either by new rules replacing old ones

38 

 Property Rights, Governance, and Power Balances

or by new rules serving as a supplement to the old rules. The latter concern forms of reinterpretation of existing rules, changing the impact of institutions even if their formal setup remains unchanged. In practice, these mechanisms are partly overlapping. In this context, layering is in a privileged position, as a typical form of a reform processes (see van der Heijden, 2011 for a conceptual discussion). Due to both savings on transaction costs (Williamson, 1975) and to the “logic of appropriateness” (March & Olsen, 1989), reforms mostly consist in adjustment of existing institutional traits; even new policies are often developed on the basis of existing rules and routines, as March and Olsen (1989) point out regarding the emergence of the Norwegian oil sector around 1970. When displacement is put into effect in revolutionary situations, implementation will take place over time and is maybe not completely realized. If so, the intermediate period will be one of layering, with the possible result of tensions between the old and new elements of the institution. Or, a process of displacement may go beyond a tipping point where its basic mode of operation is clearly changed, to the effect that some of the old models are retained. In both examples displacement is a special case of layering, understood as a relatively stable combination of old and new elements. When old and new rules coexist, the interplay between them may vary significantly. Four different types may be distinguished. (i) Reinforcement. Interaction effects reinforce one or both sets of rules; the compound serves the same goals as before, but more efficiently. Legislation on worker protection may be broadened in scope by the introduction of new technologies, and thereby the focus on already existing working environment problems is also strengthened. (ii) Watering down. Even if new rules are introduced to modify the old ones, over time the old regime turns out to be the more viable, and the new rules are removed. The reform of the Norwegian commercial banks in the late 1970s quite quickly lost its momentum and was abolished after a few years. (iii) Transposition. New rules contribute to the emergence of new bargaining situations, not expected by those who originally designed them. Reform of the pension scheme in Norway in the 2000s may serve as an example. (iv) Constant conflict. Instead of new rules complementing the old ones, persistent tensions are created, with neither old nor new rules winning in the long run. Vocational schools are examples of hybrid organizations where the goals of learning by doing stand in continuous conflict with theoretical training. Institutional change will also be shaped by tensions between power structures and aggregate action. If taken to the extreme, emphasis on aggregate action will limit itself to zero-sum games, where there are no win-win situations, or even compromises, and narrow room for the exercise of power is left. Mostly – if not necessarily – aggregate actions lead to changes by diminishing or hollowing out existing institutions but are unsuited to solve collective action problems. In that case, strengthening institutional arrangements presupposes dictatorial power to overcome Prisoners’ Dilemma situations. Positive reforms, in contrast, assume some form of institutional compromise, either by key actors coming together to prevent a Tragedy of



A Brief Note on Property Rights 

 39

the Commons or by wide-ranging agreements or legislation including the state as one of several parties. At best, such processes have more in common with Ostrom’s (2009) theory of preservation of the commons than with standard solutions of Prisoners’ Dilemmas. Relevant types of actor groups to achieve these results are of course powerful groups, such as employer associations, trade union federations, politicians and civil service bureaucrats; in addition, intellectuals and social scientists play a decisive role.

3.2 A Brief Note on Property Rights Property rights constitute a central part of democratic order and are at the same time a precondition to the operation of any type of modern economy (Fligstein, 2001). They are constituted by a bundle of rights in objects, embracing rights of exclusion, rights of use and rights of transfer (Carruthers & Ariovic, 2003). These rights are regulated by the legislation, by binding agreements or by established custom. In modern economies property rights may be exerted by juridical persons (e.g. companies buying and selling objects), by natural persons directly (e.g. personal ownership of material assets) or indirectly via shares in joint-stock companies. Shareholders, in turn, have three sets of substantive rights: the right to vote in the General Assembly (exclusion), the right to the residual income of the company (use) and the right to sell and buy shares (transferability). In a political perspective exclusion is conditional on state regulations of the right of owners to solely dispose their property; by implication there are limits to the right of owners to exclude other actors from decisions within enterprises. Property rights are endorsed by several modes of legitimacy, to a large extent independently of each other: “a principle of utility and efficiency, a principle of justice and equality, and a principle of desert based on labor” (Munzer, 2009, p. 3). The absence of one unified form of justification is an obvious source of conflict. In working life property rights were a zone of intense contention during most parts of the 20th century, when significant parts of the labour movement denied their legitimacy (Falkum, this volume). At the same time property rights vary considerably between societies, as illustrated by the differences between liberal and coordinated capitalism. From the Varieties of Capitalism perspective, however, the potential for change may be underrated, even if stability is considerable. The most important source of stability is Article 105 of the Norwegian Constitution, stating that in cases of expropriation, owners have the right to full economic compensation for their losses. Consequently, widespread nationalization of industrial property would require changes in the Norwegian Constitution; otherwise, the economic costs to the state would be prohibitive. Questions of nationalization of private property have seldom been pushed to the extremes in Norway, but the contribution of Article 105 to stability was demonstrated in the late 1970s when the

40 

 Property Rights, Governance, and Power Balances

Labour government decided to “socialize” the commercial banks (NOU, 1976, p. 52; Bergh, 1982; Engelstad, 2012a). The core of the reform was a drastic change of the composition of members of the board of representatives and board of directors by appointing political representatives to the majority of seats in these bodies. The reform was met with intense resistance and claims of compensation by the shareholders by reference to the Constitution. However, when a conservative government came to power a few years later, the bank reform was revoked. The conflict over compensation, as foreseen in the Constitution, never came to a conclusion, but in all probability the state would have run into such costs that the reform would have been reversed anyway. Simultaneously, in a society in change constitutional guarantees of property rights are open to reinterpretation, concerning both focus and extension. Changes mostly take place through indirect processes, within limits that are not established once and for all. Hence, the appropriateness and acceptability of possible changes must be tried out in each separate case. Nevertheless, a possible specification of the limits to change may be the following: Given that property rights are democratic rights, changes in their formulation and interpretation are legitimate to the extent that they broaden democracy. But, in this case as always, changes entail unanticipated consequences (Banner 2011), and changes do not come about without struggles over power and interpretation of social norms.

3.3 The Empirical Landscape: Background In order to describe and discuss changes in property rights following from the successful, and failed, reforms in working life after 1970 in Norway, it is necessary to take a glance at the Basic Agreement between the employer and the trade union confederations of 1935 (Falkum, this volume). The agreement marked the introduction of a long period of labour peace; and due to its character of a class compromise, it also prepared the ground for the extensive construction of the Norwegian (and Scandinavian) welfare state. It also was a precondition for the ensuing reforms of the industrial democracy of the 1970s. As strongly underscored by Falkum, the Basic Agreement was established by the labour market parties in opposition to state regulation of working life and labour conflict by law. Regulation by agreement allowed for a higher degree of stability and delivered a higher degree of legitimacy than what would have been obtained by legislation. The Basic Agreement is often characterized as the foundation of stable bargaining relationships, but it is something more. Even though it contained many elements included in earlier wage agreements, it represented a decisive novelty (but see Olstad, 2010 for a contrasting view), both for what it mentioned and for what it left unstated. Wages and working conditions are not the subjects of the agreement; it solely regulates the relation between the bargaining parties. It is not limited to given



Case 1: Job Related Co-Determination 

 41

industries or to a specified number of companies but embraces the whole of working life covered by the employers and trade union federations, including non-unionized employees in companies covered by the agreement. Mutual recognition was the core of Article 1 in the 1935 agreement, which stated that the employer and trade union confederation accept the other party’s “full right of organizing”. Taken together with the statement that “trade union officers within each company are accepted as representatives of workers” (Article 2), the generality of the agreement implies a decisive change in the employer–employee relationship in that the civil rights of organizing was recognized within the confines of the enterprise, within the framework of “calm and good cooperation” (Article 5). Just as significant is what remained unmentioned in the agreement. Property rights were not referred to anywhere. The managerial prerogative was present only indirectly (in contrast to the first nationwide agreement of 1907) by its limitations and by the restrictions employers have to accept within the enterprise. Although the Basic Agreement became the stepping stone to a long period of labour peace and a spirit of “good cooperation”, as expressed in Article 5, it represented a compromise, not general consensus. It thereby allowed for a new way of handling class conflict. This was not only true for the general guarantees of the position of trade union officers; it also concerned the lacunae of property rights and managerial prerogatives. Even if abolishing private property never was a serious issue in the Norwegian labour movement (Hagen, this volume), paradoxically, the agreement allowed for attempts at reforms in property rights. This springs out of the only tacit recognition of managerial prerogatives combined with even more implicit recognition of property rights. The managerial prerogative has its basis in property rights; management is acting on behalf of owner(s). By accepting management’s prerogatives, what the trade union has recognized as property rights is the rights of owners to the degree that they are actively expressed in employer prerogatives, which in turn are also conditional. Exactly where this limit is to be drawn, however, is not expressed in any agreement or legal statement; it can only be explored in practice. This is also what was tried out after debates over more than three decades after the Basic Agreement was signed and then subsequently reinforced, expanded and modified in 1966 (Falkum, this volume). Without the Basic Agreement the reforms of the 1970s on health and safety and on employee participation on boards of directors, which modified both property rights and employer prerogatives, could hardly have taken place.

3.4 Case 1: Job Related Co-Determination In most of the industrialized world legislation on worker protection emerged from the latter part of the 19th century and was further developed during the 20th century. In Norway the years of decisive changes were 1935, when a full scale Labour Protection

42 

 Property Rights, Governance, and Power Balances

Act was passed, parallel to the signature of the Basic Agreement, and 1977, when the present Working Environment Act (WEA) was adopted (NOU 2010:1). The introduction of the WEA could be seen as a clear example of layering, in the sense that the existing order was supplemented by a comprehensive set of new rules. But it makes more sense to regard the reform as a case of displacement. The new elements were not so much a supplement to the old ones; rather, the rules of the game were decisively changed, even though the goal was kept more or less intact. The changes concerned both the extension of the legislation, the organization of the protection of the working environment and the allocation of rights to the relevant parties within the enterprise. A central instance of control was moved from an external agency, the Labour Inspection Authority, to the employees themselves; thereby the authority structure in the company was affected (Kalleberg, 1993). The background to the adoption of the WEA in 1977 includes several factors. The period of reconstruction and economic austerity after the Second World War had come to an end by the mid-1960s. Nevertheless, the 1960s became a period of labour unrest, making both employer and employee organizations search for new solutions (Thorsrud & Emery, 1969). Research on job-related problems was sponsored by both labour market parties, leading into to a large-scale research project called the Cooperation Programme (“Samarbeidsforsøkene”). The programme studied experiments with semi-autonomous groups in several industries and formulated a basic list of so-called Psychological Job Requirements pertaining to autonomy at work (Thorsrud & Emery, 1969; Engelstad, Emery & Thorsrud, 1970). From an international perspective the Cooperation Programme represented a significant innovation, but it did not achieve a breakthrough in the Norwegian trade union movement until the early 1970s, when a large-scale working environment study documented the overwhelming presence of health and security problems (Karlsen et al, 1975). The innovative character of the WEA is threefold. First is the extension of legislation. The criteria of health and security damages were drastically extended. In addition to material threats, such as the presence of dangerous gases, exposure to extreme temperatures or ergonomic problems, the act defined psychosocial issues as part of the working environment, something that was to have significant consequences in the future. Second, the WEA prescribed jobs to be set up in a way that would foster the individual development of employees. Job rotation was part of this, but the focus was much wider, embracing the content of jobs in general. In its 1977 version these points were listed in Article 12 (presently Section 4) of the act, obliging employers to strengthen personal autonomy and promote on-the-job training. Third, the WEA reorganized the responsibility for improvement of the work environment by prescribing two new positions in the company: an internal ombudsman for health and security to be elected among employees and a Working Environment Committee with 50/50 representation of management and employees. The ombudsman was given the authority to bring production temporarily to a halt in case serious security



Case 1: Job Related Co-Determination 

 43

threats were disclosed, whereas the Work Environment Committee would make recommendations on the future work environment. Obviously, these changes were radical, maybe even more radical than the main actors realized at the time. In hindsight it is striking how the points of Article 12 reflect – probably unintentionally – the concept of alienation from work in the early writings by Karl Marx (1844). A few years after the adoption of the WEA, the organizational part of the reform was characterized as not only a new form of job design but a significant example of participatory democratization (Kalleberg, 1983). This leads to a broader reflection on the issue of institutional change. What is the institution at stake, and in which ways is it changed? Above all, the WEA represents a decisive change in labour–management relations, as the roles of both employer and employee are substantively changed. By transferring decisions formerly made by management alone to the Health and Safety ombudsman as well as to the Working Environment Committee, authority relations within the enterprise are redefined. These changes in roles are reinforced by the job quality prescriptions represented by the original Article 12. Moreover, employers are legally obliged to actively improve the working environment. Given that employer prerogatives were substantively modified, it may be argued that property rights also were affected, albeit indirectly. From the view that shareholders are owners of the enterprise in a strong sense, with the board and management acting as their sole delegates (Jensen & Meckling, 1976), it follows that the WEA is an infringement on property rights. By taking the perspective of stakeholder theory, holding that what shareholders own is not the company but merely shares entitled to certain rights (Blair, 1995), the conclusion becomes less drastic, but it still makes sense that property rights are affected. The debate between these two positions has a parallel in contemporary debates on conceptions of management in Norwegian working life. Two competing modes of thinking were juxtaposed in the late 1960s and early 1970s, represented respectably by George Kenning and Einar Thorsrud (Kalleberg, 1991; Qvale, 1995; Sørhaug, 1996), and the contrast between them became a dominant issue for decades. Kenning, representing a Scientific Management approach, held that management was a specific profession and should not grow out of expertise in the technical aspects of production. Thorsrud, on the other hand, in line with the Cooperation Programme, regarded management most of all as a sort of facilitating task, serving the enterprise and its “inhabitants” as a whole. Thorsrud’s idea of leadership gained considerable support and became a decisive element in the idea of a specific Norwegian model of management, even though the Kenning tradition kept a stronghold in the core of Norwegian manufacturing. This division was one reason why the idea of semi-autonomous work groups did not really take off in Norway. Trade unions did not include semi-autonomous work groups among their main issues. Their gains from the reform were mainly located elsewhere. Within the enterprises the WEA increased the power of unions considerably by opening a much

44 

 Property Rights, Governance, and Power Balances

wider scope for bargaining with management. In addition to the opportunity to fill the positions of ombudsman and members of the work environment committees with their members and officers, their political role was strengthened. In an alternative context, problems in the working environment might be solved by agreements and negotiations between the labour market parties, either decentralized to enterprises or between employer and employee federations. As a consequence of the legislation changing the rules of the game and broadening work environment problems from protection of physical health to a wide range of issues in employment relations, debates and conflicts were brought into the political–public sphere and transformed into significant political topics. Thereby the relative strength of unions in relation to employers was considerably increased. If the position of trade unions was strengthened, it was to a large extent accepted by employers, as documented by the Leadership Study 2001 (Gulbrandsen et  al., 2002, Chapter 8). One obvious reason is the political climate and the hegemonic position of the Trade Union Federation (LO), not least based in the Basic Agreement. This may explain the introduction of the reform but not its subsequent acceptance. A second set of reasons is found in the productivity gains by a good working environment. In modern working life, employee empowerment and participation are necessary elements. Finally, the law reduces the risks to the enterprises of being held responsible for serious health damages, with the possibility of being taken to court and suffer serious economic losses. When rules are codified in legislation, employers are discouraged from the excessive cutting of costs and thereby from a race to the bottom threatening the productivity effects of a good work environment. The role of intellectuals in the process should not be underrated. The Cooperation Programme became a pivotal element in driving the reform process forward. Researchers linked to the project later played a key role in formulating the revised and extended version of the WEA in the mid-1970s. One aspect of the research programme, that of psychological job requirements, thereby enjoyed the success of being implemented in Article 12. The part concerning semi-autonomous work groups, however, never gained broad acceptance in Norway, neither at the time of the reform nor later on. An additional reason for this failure may be sought in its relationship to employer prerogatives. If specific limitations on employer prerogatives are codified in legislation or in basic agreements, they may acquire stability in the long run. Semi-autonomous work groups are a different matter; they are part of the organizational structure and are solely dependent on the discretion and decisions exerted by employers. In periods of continuous reorganization of work processes, such arrangements are necessarily precarious. This is illustrated by experiences in Sweden, where experiments with worker autonomy in the car industry seemed to be successful for a period; nevertheless, most were closed down by the end of the 1980s. The stability of institutional change in the working environment rests on the common interests of employers and employees, solidified by legislation. As indicated, it has elements both of displacement and of layering, depending on the analytical



Case 2: Employee Participation in Top-Level Decision-Making 

 45

perspective. The effects of working environment issues entering the political sphere might be seen as a special form of conversion; in the description by Mahoney & Thelen (2009, p. 17) “… rules remain the same but are interpreted and enacted in different ways”. Implied in this description is that the arena remains the same. As this is not the case, it makes more sense to point to the special form of layering characterized above as transposition. Even if the issues remain more or less unchanged, they become subject to different power constellations in the political sphere rather than in local bargaining, and are thereby handled within different horizons of interpretation.

3.5 Case 2: Employee Participation in Top-Level Decision-Making By the early 1950s the Labour Party, then in power, introduced the idea of employee representation in the boards of directors in joint-stock companies. An early attempt was made in some state-owned companies (Bergh, 1983, p. 104; Falkum, this volume); the long-term goal was the extension of industrial democracy to the private sector. In subsequent debates from the early 1960s and on, both employers and trade unions expressed a preference for democratization via agreements rather than by legislation. However, the process came to a standstill due to a combination of complexity of the issue and the parties’ unwillingness to give up established positions. The initiative could thereby move to the political sphere. In the late 1960s a governmental committee was appointed (the Eckhoff Committee), with the mandate of presenting alternative forms of economic democracy. The committee had representatives from the Trade Union Federation, the Employer Association and an independent group of legal scholars. When the committee presented its conclusions, the minority of members representing the LO recommended the formation of corporate assemblies with the task of electing the board of directors and overseeing the general strategy of the company. The majority, including the employer representatives, proposed instead that employees should be represented in the board of directors (Eckhoff Committee, 1971; Bergh 1983, p. 114). When the Labour Party came to power shortly afterwards, a revision of the Limited Companies Act containing both alternatives was presented and later adopted by Parliament. The revised act opened the possibility for employees to demand representation in up to one-third of the seats on the boards of directors in companies over a certain size (at the time 50 employees). Moreover, the creation of corporate assemblies was prescribed in firms with more than 200 employees. In the following years employee representation on the board was extended in several respects, into parts of the public sector and in the relationship between parent companies and daughter companies. The process was crowned in 1980 by an amendment to Article 110 of the Constitution: “Specific provisions concerning the right of employees to co-determination at their workplace are laid down by law”. The article does not prescribe given arrangements but states unambiguously that legal

46 

 Property Rights, Governance, and Power Balances

regulations of co-determination – and thereby of property rights – is not in conflict with constitutional guarantees against expropriation in Article 105. Obviously, from a shareholder perspective employee representation on the board of directors is a significant intervention into property rights; shareholders are deprived of the full right to elect representatives to look after their financial interests. Hence it is a major case of institutional change. How could this come about, and later be accepted by employers? Here, too, the spirit of cooperation and trust between employers and trade unions rooted in the Basic Agreement is a salient background factor. Another is Labour Party hegemony developed over almost three decades in power. The party had a double position: on the one hand securing a broad popular base for modernization of Norwegian society, and on the other hand acting as a guarantee against political extremism. On this background a complex process played out. The main interest of the trade unions lay in the introduction of corporate assemblies (Hagen, this volume). These were seen as possible tools for broadening employee participation in decision-making at a later stage. The employer representatives of the Eckhoff Committee were strongly opposed to this new body, partly for the same reasons. However, they could hardly block every possible change in the power relations; that would come close to destroying the cooperative relationship with the trade unions. It should also be underscored that both parties saw themselves (but not necessarily the other) as moving within the limits of constitutional protection of private property. Hence, the employer side reluctantly opted for employee representation on the boards as their main concession to the struggle for industrial democracy. At least three arguments were in favour of this move. First, the Conservative Party, in support of general “ownership democracy” had introduced into their programme that employers should be encouraged to invite employees onto the boards if it was deemed appropriate (Bergh, 1983, p. 110). Second, the day-to-day operations of companies were not affected and board representation did not change employer prerogatives and concerned only a minority of board members. Third, a study on property rights by the legal doyen Professor Johs Andenæs concluded that a minority representation on the boards would not be a break with constitutional guarantees (Andenæs, 1970). Fourth, a strategic consideration: the introduction of a corporate assembly would imply a new body whose mode of operation was unsettled, whereas the board of directors was a well-established institution where the business elite felt at home. In the following years, this meant that the employer side accepted board representation, whereas they were still arguing against the corporate assembly (Hagen, this volume). The undermining of the assembly took off already in the first years after the Limited Companies Act was revised (Engelstad & Qvale, 1977) and led to a massive decline in its prevalence after a later revision of the act in 1988 (Gravdal, 2010; Hagen, 2014). The position of intellectuals in the process is interesting. One might assume that the intellectuals would be in favour of the reform, but actually they were not. The researchers running the Cooperation Programme of self-governed work groups were



Case 2: Employee Participation in Top-Level Decision-Making 

 47

opposed to the board reform because they felt that democratization was introduced in the wrong place. Their concern was enhanced autonomy among workers on the shop floor, in line with their own research, and workers’ entrance into the board room would hardly have effects in that direction. However, with the adoption of the WEA a few years later their concerns were met. A radical faction in the Labour Party argued in favour of “full industrial democracy” (Anker Ording, 1964, Chapter 3), much in line with experiments of worker participation in Israel and later in Yugoslavia. Another group of sceptics was found among Norwegian sociologists (Bergh, 1983, p. 116). Their criticism was directed against assumed difficulties for employees to participate in the deliberations on the board. Their fear was co-optation of employees’ representatives, eloquently presented in the new theory of “model power” (Bråten, 1973). Later research concluded that the fear of co-optation to a large extent was unfounded (Hagen, 2010). In the three decades after 1980, two significant events have taken place (Engelstad, 2012a; Hagen, this volume). First, a committee jointly appointed by the LO and the Labour Party, the Skytøen Committee, recommended the extension of the level of employee representations on boards from one-third to one-half (strengthening of the Company Assembly was not proposed). Again intense protests from the employer side ensued. In the end the result was modest, seen from the perspective of the committee. In the ensuing revision of the Limited Companies Act in 1988 the existing proportion of employee representatives on the boards was left untouched; the only change was that the limit for board representation was lowered from 50 to 30 employees. Since then, no forceful proposal to extend representation has been made. Second, the 1988 revision allowed companies to dismantle the corporate assembly in exchange for somewhat stronger employee representation on the board. The result is that corporate assemblies have practically disappeared, except in a few publicly owned enterprises (Gravdahl, 2010). Thus, board representation is established as the one relevant modes of co-determination on the enterprise level in Norwegian working life. Over time this arrangement has become strongly supported by both parties in the labour market and by the general public (Gulbrandsen et al., 2002, Chapter 8). One possible reason for the acceptance of board representation among employers is the debate over and subsequent rejection of the extension to 50 per cent representation. It was made clear that employers did not face the risk of losing control over decision processes on the boards. Moreover, the company assemblies did not turn out to be the efficient political tools initially envisaged by representatives of the LO (Hagen, this volume). From a long-term perspective none of the parties realized their preferred option and none could declare that they got their way. Employee representation on the boards turned out to be a second-best solution for both, pre-empted by none of them but a compromise they could live with. The question remains whether employee representation will be stable in the future. Despite the positive attitudes of the business elite, it might plausibly be expected that employers in the long run would resist representation in general (for

48 

 Property Rights, Governance, and Power Balances

an example, see Nicolaisen and Trygstad, this volume). In some industries, like retail trade, where large chains are broken up into franchise units, this may actually be the case – likewise in parts of the “new economy”. At the same time, there are reasons to expect that the system will mostly remain intact. Employee representation implies an open channel of information on the shop floor, which is advantageous to a large part of board directors. Moreover, the arrangement is a source of trust in the management/ employee relationship. Both of these factors can be expected to have positive effects on productivity. Finally, the existence of employee representatives is a source of legitimacy for the company vis-à-vis the environment if, for instance, the board is to make controversial decisions, such as in downsizing or even closing the company. When employee directors also stand behind such decisions, it becomes easier to gain acceptance in the public (Hagen, 2010). Regarding the processes in the light of mechanisms of institutional change, the following picture emerges: The initial change in co-determination is a case of layering. The Limited Companies Act existing up until 1972 was partly changed when the new elements of employee co-determination by means of corporate assemblies and employee elected members on the boards of directors were incorporated. But the original version of institutional layering did not work out as intended. Whereas the company assembly turned out to be dysfunctional, employee representation showed itself to be stable and was even slightly reinforced after a process of delimitation of its extension. If, on the contrary, the company assembly had been the viable of the two, it would have made sense to characterize the reform as a case of displacement; then the basic rules of operation would have been changed, as they were in the working environment reform. But this did not happen; the result was a rather watered down form of layering.

3.6 Case 3: Gender Quotas to Boards of Directors Albeit related, the third case to be discussed is of a different type. Gender quotas to boards of directors in private companies were introduced by legislation in Norway in 2003, to be fully enacted from the beginning of 2008. The background to this reform is drawn up by Teigen (this volume). After a revision of the Gender Equity Act in 1981, state initiated committees and councils were prescribed to have at least 40 per cent of both genders as members, the same was true for state-owned agencies and enterprises. However, during the 1980s and 1990s a wave of conversion of state-owned enterprises into listed firms was taking place; as an unintended side effect the recently privatized companies were free from the requirement of gender quotas. Hence, the equal opportunity issue was facing a backlash, which was experienced as intolerable by key actors, given the strong position of gender equity policies established in the public sector from the early 1980s (Hernes, 1987). One way of compensating this loss could have been to introduce gender quotas in these newly privatized companies.



Case 3: Gender Quotas to Boards of Directors 

 49

This, however, would create an inconsistency between these companies and the large number of firms already in operation. What presented itself as an option for cutting the Gordian knot was broadening the focus and introducing gender quotas to boards of directors in the private sector in general. Reform initiatives were introduced in several steps. The first step was made in 1999 by the Centre government, which proposed to introduce gender quotas in all joint-stock companies. This very radical reform proposal quickly fell to the ground. A shift of government to Labour brought forth a more moderate proposal, limiting the gender quotas to public limited liability companies (“allmennaksjeselskap”), which mostly included firms listed on the Oslo Stock Exchange and enterprises in the finance sector. This proposal was appropriated with a few adjustments by the ensuing centreright government. When the act was adopted by Parliament in 2003, it contained the peculiar provision that it would not be enacted if the companies themselves were able to raise the level of female board members to 40 per cent. Three years later the rate of female board members had increased to some 15 per cent, a far cry form the target of 40, and the law came into operation. The main employer association, the NHO (Confederation of Norwegian Enterprises), was predictably opposed to the act. Even though official declarations from employers expressed very positive attitudes to gender equality, they were resistant to legislation to reach the goal (Ot.prp. 97, 2002-2003). The NHO had taken initiatives to raise women’s qualification for board positions, but they were far from sufficient in reaching the goal. Consequently, employers held a weak position in their resistance to the new act, having demonstrated their inability to reach a goal they supported. As the LO was quite indifferent to the reform, this time there was no option to forge an alliance against the state. Advocates of the reform were located in two circles: first, feminist activists, among them female scholars with close links to politics, and second, parts of the civil service and the political establishment. The centre-right government was divided on gender quotas, which obviously was at odds with the views in the liberalist fraction. The reform won through an internal coalition of the Christian Democrats with some conservatives, backed by the gender equality policies within the state sector with a stronghold in the Ministry for Family and Gender Equality Affairs (Teigen, this volume). In retrospect the quota reform has proved to be very successful, the target of 40 per cent was soon reached, contrary to wide-spread assumptions in business circles that it would not materialize due to lack of competent women. This scepticism may also have contributed to the relatively modest resistance among top business officials. Protest never materialized in anything like sabotage or threats of legal prosecution. Another reason for the success may be found in the very strict sanctioning on breaches, which was laid down in the Limited Companies Act long before the quota reform. By introducing the reform into the current act, the existing sanctions followed

50 

 Property Rights, Governance, and Power Balances

automatically. Hence, the quotas could not be delegitimized by the very strong claims to compliance. The gender quota reform is a clear-cut case of layering, without the tensions and ambivalences discussed above. The institutional implications may be compared to employee representation on boards. The first (and drastic) proposal of the reform referred explicitly to the board reform of 1972 to legitimize the new intervention (Teigen, this volume), to some extent with good reason as the quota reform obviously constituted a limitation of property rights in selecting board members. Simultaneously, the new reform had a quite different orientation, something that may have contributed to its later acceptance. Most importantly, female members of the board were not expected to represent specific group interests in opposition to shareholders or advance some form of class confrontation, as might be the case with worker representatives. Moreover, the number of companies covered by the law was much more restricted, comprising between 300 and 500 companies, less than 10 per cent of the number of enterprises in which employers have the right to claim board representation. Another parallel to board representation is worth reflection. As mentioned, the proposal of the Skytøen Committee to extend employee board representation was rejected after intense debate. A similar debate took place in the aftermath of the quota reform. Representatives of the centre-left government in power up to 2013 launched the possibility of extending the reform from listed firms to large private limited liability companies (“aksjeselskap”), but a formal proposition never came on the agenda. In the public debate the idea once more met resistance among employers. The withdrawal on the part of the government may be seen on the background of this resistance, but additionally both normative and practical arguments may be relevant. Shareholding in listed firms has a more neutral and less personal character than ownership in family-owned firms, even if they are large. Hence, state intervention in board composition in the latter becomes more problematic. On a practical level, problems arise in delimiting the relevant target group. How is a “large” company to be defined, and how is the delimitation to be defended? Another problem resides in the relationship between mother and daughter companies, where mothers often are listed while daughters are unlisted companies. Commonly, boards of daughter companies are manned by the top officials of the mother company; interfering with this wellestablished arrangement entails the risk of intervening into the appointments to the group of top executives. The exact reasons for withdrawing the proposal cannot be known, but the arguments cited indicate that extending the quota reform at least comes close to transgressing the legitimacy of state intervention. Whether the reform in its present form will remain stable is another question. The present centre-right government has not expressed wishes to change or abolish the law in the form it was adopted in 2003. On the other hand, employers might search to counteract the law, but their motivation is probably weak. The law does not infringe decisively on the power of decision-making among employers; it only sets a moderate



Concluding on Mechanisms and Power Relations 

 51

limitation on the selection of board members, widely different from, for example, the political appointment of board members in commercial banks for some years around 1980, mentioned above. Employer attitudes towards the law might also be influenced by productivity effects, assumed to be positive for employee co-determination, both on the shop floor and in the boardroom. Along a similar line the proposal of the bill on gender quotas also argued that it would lead to increased productivity (Ot.prp. 97, 2002-2003). There is no sign that this is the case. Nor are there serious signs of decreased productivity (Dale-Olsen, Schøne & Verner, 2012; Engelstad, 2012b), despite claims to the contrary (Ahern & Dittmar, 2012). Hence, the most probable scenario is that the reform will remain stable for the foreseeable future.

3.7 Concluding on Mechanisms and Power Relations The introduction pointed out that among mechanisms of institutional change layering holds a privileged position; simultaneously the combination of old and new elements may follow different paths, as illustrated by the three cases presented. The introduction of gender quotas is a case of reinforcement. One reason for its success was that it was followed up by very strict sanctions. These sanctions, however, were already in existence for any breach with the regulations of the Limited Companies Act. On principled grounds it could be argued that this is a case of constant conflict rather than reinforcement, to the extent that gender quotas constitute a substantive challenge to property rights. The counter argument, which has gained support in practice, is however that this is not a breach of property rights, but rather a specification. The Working Environment legislation of 1977 represents a mix of layering and displacement. It was argued that it comprised more traits of displacement than of layering, as the rules of the game were changed when the enactment of the law shifted from the Labour Inspection to agencies and positions internal in the companies. A secondary effect emerged because the reform gave increased power to trade unions. Moreover, in a process of transposition, power games were extended from the enterprise to political debate in the public sphere. In its original version, employee participation in top-level decision-making was also a form of layering with clear elements of displacement. If the conception of the company assembly had been upheld and reinforced, it might be relevant to talk of displacement in this case, too. Albeit on limited issues, the company assembly was conceived as the top body of the company. In practice it never worked that way – the company assembly gradually vanished. The reform thereby took a clear-cut form of layering. At the same time it was watered down, as the board remained the only relevant arena for participation. Mechanisms of change are shaped by the actions of powerful actors. The most common theory of power in studies of institutional change is the power-resource

52 

 Property Rights, Governance, and Power Balances

theory, which measured the strength of the labour movement by the organizational strength of trade unions and its support by political parties of the left. These are obviously important factors, but a weakness is that they focus mainly on the actors, to some extent disregarding their complex relationships and the normative restrictions they are subject to. Particular for the stories presented here is the relationship between actors – employers, trade unions and the state – who realize that they are bound together in a commonality of interest. In some situations employers and trade unions are bargaining in the shadow of the law; in others trade unions or employers may side with the state against the third party. The Basic Agreement would never have come about without the threat from the state of regulating by legislation – an alternative the both parties found worse than coming to terms with each other. On the other hand, even if both parties wished to come to an agreement on employee co-determination, it would never have come about without active intervention via legislation. The obvious reason is that questions of property rights would be involved, an issue the parties were unable to handle on their own. This had already been made clear when the Basic Agreement avoided these questions. As a side effect this also created an opening for the labour movement to legitimately press for restrictions of property rights, albeit without the power of reaching a conclusion. Similarly, the new health and security arrangements foreseen by the WEA could hardly be the result of bargaining between the parties, as it entailed a limitation of employer prerogatives. In both cases employers could be regarded as the victims of the coalition between trade unions and the state – in both situations a state manned by a Labour government. And the employers were bound by the Basic Agreement and the necessity to come to terms with the unions. However, they were far from powerless. In these cases their main power did not rest in conventional sources of conflict (lockouts, capital withdrawal, etc.) but in their ability to encounter state intervention by mobilizing arguments of a normative character – constitutional rights and substantive considerations of property rights. Additionally they could argue on the basis of efficiency linked to the organizational setup of enterprises: What are the conditions for a workable board of directors? What constitutes the most basic interests of employees? One weakness in the power resource theory is its disregard of the power anchored in norms and ideas (Korpi, 1985). But norms and ideas, as they are exposed by intellectuals and/or social movements, obviously have had an impact on institutional change in addition to power resources based in membership or in political positions. This was obviously the case in the working environment reforms, where ideas of academic researchers played a significant role in two steps: first by launching the Cooperation Programme and the conception of psycho-social job requirements in the 1960s and then by mapping the working environment problems in the early 1970s. They gave decisive inspiration both to the arrangements of job-level participation and to the formulation of Article 12 in the WEA, something hardly possible in bargaining through “hard” power resources alone. Gender quotas represent a similar case. Even if



Concluding on Mechanisms and Power Relations 

 53

the feminist movement in a broad sense played a relatively passive role in the debate, the values of feminism were overwhelmingly present and strongly represented by a coalition of central activists, bureaucrats and politicians of the opposition. Here, too, the force of “hard” power would have been insufficient to bring about the result. The contrast case is industrial democracy and board representation. Here, intellectuals turned out to be the losers. Not so much due to their general values, but to their insufficient specification of the organizational problem. Their critique of board representation emanated from two sides: a first choice of transferring overarching decision-making to an assembly where shareholders would form a minority and, second, the fear of co-optation of worker representatives in board meetings. The former argumentation remained underspecified and thereby lost its persuasive power; the latter was maybe overspecified by a focus on one single aspect. Common to both was disregard of property rights. A weak understanding of property rights may have been true for job participation activists and feminist activists as well, but in both cases their possible neglect was tempered by the necessary operationalization in the process of legislation. The intellectuals promoting wide-ranging industrial democracy stayed aloof from processes of legislation. After the defeat of the Skytøen Committee, the issue remained settled for a long time. Given that property rights are democratic norms, this leads up to a brief reflection on the norms of democracy in institutional change. If bargaining hardly takes place without normative considerations (Elster, 1989), normative considerations become more powerful when bargaining implies interpretation and argumentation about norms of democracy. When the joint action of the government and trade unions pressed for the representation reform, the argument of democratization was difficult to resist. When later the employers were able to put pressure on the corporate assembly, this was also dependent on the perception that democratic aspects of the original reform were not displaced. The introduction of the WEA took a similar form. In this case, in addition to the labour market parties, intellectuals in particular represented democratic norms. From the early 1970s to the turn of the millennium general conceptions of democracy changed. Participation on issues of worker welfare was the main democratic aspect of the workplace reforms. The democratic content of the gender quotas was broader and narrower at the same time: narrow in the sense that the target group of the reform is more diffuse – women as a group – whereas the underlying goal of equal opportunity is broader. It is a norm that does not spring out of group interests but out of a generalized notion of democratic inclusion. It does not require much ingenuity to imagine that this will also be a driving force in institutional changes in the future.

54 

 References

References Ahern, K. & Dittmar, A.K. (2012). The Changing of the Boards: The Impact on Firm Valuation of Mandated Female Board Representation. Quarterly Journal of Economics, 127, 137-197. Andenæs, J. (1970). Bedriftsdemokratiet, eiendomsretten og grunnloven. Lov og Rett, 113-138. Anker-Ording, Aa. (1964). Bedriftsdemokrati. Eiendomsretten og Grunnloven. Oslo: Universitetsforlaget. Banner, S. (2011). American Property. A History of How, Why and What We Own. Cambridge: Harvard University Press. Bergh, T. (1982). Bank og demokrati. In Francis Sejersted (Ed.), En storbank i blandingsøkonomien, Den norske Creditbank 1957-1982. Oslo: Gyldendal. Bergh, T. (1983). Medbestemmelse eller opposisjon? Noen hovedtrekk ved den norske debatten siden 1945 om demokrati i arbeidslivet. In T. Bergh (Ed.), Deltakerdemokratiet. Oslo: Universitetsforlaget. Blair, M. (1995). Ownership and Control. Rethinking Corporate Governance for the 21st Century. Washington D.C.: Brookings Institution. Bråten, S. (1973). Model Monopoly and Communication: Systems Theoretical Notes on Democratization. Acta Sociologica, 16, 98-107. Carruthers, B. & Ariovic, L. (2003). The Sociology of Property Rights. Annual Review of Sociology, 30, 23-46. Dale-Olsen, H., Schøne, P. & Verner, M. (2012). Women on Boards of Directors and Firm Performance: Evidence from Denmark and Norway. Comparative Social Research, 29, 211-234. Due, J & Madsen, J.S. (2010). Septermberforliget og udviklingen af den danske aftalemodel. In T. Bergh (Ed.), Avtalt spill. 75 år Hovedavtalen LO-NHO. Oslo: LO og NHO. Eckhoff Committee (1975). Innstilling om demokrati i bedriftslivet. Oslo: Statsministerens kontor. Elster, J. (1989). Cement of Society. Cambridge: Cambridge University Press. Engelstad, F. (2011). Kan kjønnskvotering i næringslivet forsvares? Tidsskrift for kjønnsforskning, 35, 120-140. Engelstad, F. (2012a). Limits to State Intervention: Aspects of Property Rights in Social Democratic Societies. Comparative Social Research, 29, 235-265. Engelstad, F. (2012b). Likestilling og lønnsomhet. Hva vet vi om økonomiske virkninger av kjønnskvotering til ASA-styrer? Søkelys på arbeidslivet, 29, 258-274. Engelstad, P., Emery, F. & Thorsrud (1970). Teknologi og sosial forandring på arbeidsplassen. Et eksperiment i industrielt demokrati. Oslo: Tanum. Engelstad, P. & Qvale, T. (1977). Innsyn og innflytelse i styre og bedriftsforsamling. Oslo: Tiden. Fligstein, N. (2001). The Architecture of Markets. Princeton: Princeton University Press. Gravdal, Andreas K. (2010). Bedriftsforsamlinger. MA Thesis, Department of Sociology and Human Geography, University of Oslo. Gulbrandsen, T., Engelstad, F., Klausen, T.B., Skjeie, H., Teigen, M. & Østerud, Ø. (2002). Norske makteliter. Oslo: Gyldendal Akademisk. Hagen, I.M. (2010). Det mektige mindretallet. Ansatterepresentasjon i styret mellom Corporate Governance og Industrial Relations. PhD dissertation, University of Oslo. Oslo: Fafo report 2010:02. Hagen, I.M. (2014). Bedriftsforsamlingen – pleie eller plikt? Fafo-notat 2014:03. Oslo: Fafo. Hernes, H. (1987). Welfare State and Woman Power: Essays in State Feminism. Oslo: Norwegian University Press. International Property Rights Index (2014). http://internationalpropertyrightsindex.org/ Jensen, M.C. & Meckling, W.H. (1976). Theory of the firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics, 3, 305-360.

References 

 55

Kalleberg, R. (1993). Implementing Work-Environment Reforms in Norway: The Interplay between Leadership, Labour, and Law. In W. Lafferty & E. Rosenstein (Eds.), International Handbook of Participation in Organizations, Volume III. Oxford: Oxford University Press. Kalleberg, R. (1991). Kenning-tradisjonen i norsk ledelse. Nytt norsk tidsskrift, 10, 218-245. Karlsen, J.E., Næss, R., Ryste, Ø., Seierstad, S. & Sørensen, B.Aa. (1975). Arbeidsmiljø og vernearbeid. Oslo: Tanum – Norli. Korpi, W. (1983). The Democratic Class Struggle. London: Routledge & Kegan Paul. Korpi, W. (1985). Power Resources Approach vs. Action and Conflict: On Causal and Intentional Explanations in the Study of Power. Sociological Theory 1985, 31-45. Korpi, W. (1989). Power, Politics, and State Autonomy in the Development of Social Citizenship. American Sociological Review, 54, 309-328. Lundh, C. (2010). Saltsjöbadsavtalet och arbetsfreden. In T. Bergh (Ed.), Avtalt spill. 75 år Hovedavtalen LO-NHO. Oslo: LO og NHO. Mahoney, J. & Thelen, K. (2010). A Theory of Gradual Institutional Change. In J. Mahoney & K. Thelen (Eds.), Explaining Institutional Change. Cambridge: Cambridge University Press. Martin, C.J. & Swank, D. (2012). The Political Construction of Business Interests. Cambridge: Cambridge University Press. March, J. & Olsen, J.P. (1989). Rediscovering Institutions.The Organizational Basis of Politics. New York: The Free Press. Marx, K. (1844 [2001]). Economic and Philospohical Manuscripts of 1844. New York: Prometheus Books. Munzer, S. (2009). A Theory of Property. Cambridge: Cambridge University Press. NOU 1976:52. Om demokratisering av forretningsbankene. NOU 2010:1, Medvirkning og medbestemmelse i arbeidslivet. Olstad, F. (2010). Veien fram til Hovedavtalen 1899-1935. In T. Bergh (Ed.), Avtalt spill. 75 år Hovedavtalen LO-NHO. Oslo: LO og NHO. Ostrom E. (2009). A General Framework for Analyzing Sustainability of Social-Ecological Systems. Science, 325(5939), 419-422. Odelstingsproposisjon 97 (2002–2003). Om lov om endringer i lov 13. juni 1997 nr. 44 om aksjeselskaper, lov 13. juni 1997 nr. 45 om allmennaksjeselskaper og i enkelte andre lover (likestilling i styrer i statsaksjeselskaper, statsforetak, allmennaksjeselskaper mv.). Pierson, P. (2004). Politics in Time. History, Institutions, and Social Analysis. Princeton: Princeton University Press. Qvale, T. (1995). Ledelse - fra administrasjon til omstilling. In D. Olberg (Ed.), Endringer i arbeidslivets organisering. Oslo: FAFO. Sejersted, F. (1991). Demokratisk kapitalisme. Oslo: Universitetsforlaget. Streek, W. & Thelen, K. (2005). Institutional Change in Advanced Political Economies. In W. Streeck & K. Thelen (Eds.), Beyond Continuity: Institutional Change in Advanced Political Economies. Oxford: Oxford University Press. Sørhaug, T. (1996). Om ledelse. Oslo: Universitetsforlaget. Thelen, K. (1999). Historical Institutionalism in Comparative Politics. Annual Review of Political Science. 2, 369-404. Thelen, K. (2004). How Institutions Evolve. Cambridge: Cambridge University Press. Thelen, K. (2012). Varieties of Capitalism: Trajectories of Liberalization and the New Politics of Solidarity. European University Institute, Max Weber Lecture No. 2012/03. Thelen, K. (2014). Varieties of Liberalization and the New Politics of Social Solidarity. Cambridge: Cambridge University Press. Thorsrud, E. & Emery, F. (1969). Mot en ny bedriftsorganisasjon: eksperimenter i industrielt demokrati. Oslo: Tanum. Van der Heijden, J. (2011). Institutional Layering: A Review of the Use of the Concept. Politics, 31, 9-18. Williamson, O. (1975). Markets and Hierarchies. New York: The Free Press.

Eivind Falkum

4 Institutionalization and Dynamic Change of Institutions – the Basic Agreement and Tripartite Structures in Norway Basic agreements between employers’ associations and unions have regulatory impacts on work life and labour relations in the Scandinavian countries (Bergh, 2010). A basic agreement is a settlement between the labour market parties. It outlines the rights and obligations in negotiations and principles to handle conflicts of interests. In work life, agreements and legal acts are different ways to define industrial relations. An agreement embraces those who have signed it, while a law applies to all. Basic agreements and national and international legislation embed the practice of industrial relations. The Basic Agreements and the implicit strong collaboration between the social partners on production and distribution of wealth are core elements in the Nordic model, especially in Norway and Sweden (Sejersted, 2011; Dølvik et al., 2007; Moene, 2007; Bergh, 2010). The Norwegian Basic Agreement is a core element in the formation and the institutionalization of political, economic and welfare state arrangements discussed in this book. This chapter sets existing historical evidence, analysis and debates about the formation and development of the Basic Agreement into an institutional perspective with two intentions: first, to explain the emergence and design of the Basic Agreement in 1935 and second, to explain the redesign of this original set of rules in 1966 as a process of institutional change. Was the 1935 agreement a new institutional design or a sequence in a long-term process of institutionalization? How was the distribution of power in society affected by the agreement of 1935 and the institutional development that followed? The scope is how and why institutions come about and how they develop and deal with shifting contextual frameworks and power balances over time. By studying the establishment of the basic agreement and the change processes that followed until 1970, the intention is to show how positive feedback, path dependencies and institutional resilience (Pierson, 2004) interconnect in the designing and development of work life institutions. The Basic Agreement in Norway started out as a set of agreed upon rights and obligations that bound employers and employees to regulated behaviour. The basic formal rights and principles are still the same, even though the social partners since then have extended the rules in the agreement significantly. Simultaneously the effects of the agreement have expanded due to dramatic changes of context by political and economic institutions. Employment rates are about 74 per cent of men as well as women. This a very high rate compared to most other countries. The amount of employees with tertiary education has now increased and reached about 35 per cent among those 30 years old. The content of work, tasks and relations has been changed © 2015 Eivind Falkum This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Theoretical Perspectives on Institutional Design, Development and Resilience 

 57

by technologies in almost all branches. Businesses, enterprises and industries have diversified. Employment has decreased in industrial manufacturing, but has grown in public and private services. Many new institutions have developed since the 1940s. Hence, the Basic Agreement and its institutional structure now function in a much larger network and context. What are the consequences of changing contexts for institutional development? How can the formal rules and principles be the same, but still work in a completely new setting?

4.1 Theoretical Perspectives on Institutional Design, Development and Resilience The discussion draws on Paul Pierson’s (2004) analysis of path dependencies and institutional development. I discuss his concepts of positive feedback, institutional development, irreversible change, resilience, veto points and asset specificity in the light of case findings. The analysis relates as well to Mahoney and Thelen’s (2010) discussion of incremental change of institutions, including the concepts of displacement of institutional arrangements, layering of institutional elements, drift of functions due to environmental development and conversion of institutional functioning despite stable rules, values and norms. Pierson (2004, pp. 105-107) distinguishes between actor-centred functionalism and societal functionalism in the social sciences. The former rests on the idea of rational design of institutions in order to serve the interests and intentions of those who created it (p. 105). Societal functionalists, on the other hand, explain the existence of institutions by their effective responses, intended or not, to some societal problems. There are certainly examples of individual actors that have influenced the design of political and economic institutions to great extents, like building national constitutions, welfare state arrangements, universities, religions or monetary systems. These actors formulate, promote and act to establish rules, norms and understandings of realities and structures to carry them. Intentions legitimize such institutions. However, institutionalization may as well be the result of incidents and experiences that have no specific “designer”. Institutions may grow from collective processes with multiple actors, whose intentions, purposes and goals may be ambiguous, vague, tacit or simply forgotten, but the institutions may still determine human actions and behaviour. Some institutions function like perfect rational organizations while others function like nothing more than shared ideas and understandings. Rational choice models can explain the design and functions of the first kind, while the latter calls for different analytical tools. Pierson (2004), as well as Hall (2010) suggest historical institutionalism to bring appropriate theoretical and methodological perspectives for analysis of institutional change and development of institutions. Pierson (2004) makes two observations that may seem contradictory at first glance. On one hand, he develops a theoretical concept of positive feedback as a

58 

 Institutionalization and Dynamic Change of Institutions

mechanism that makes institutions find new paths to follow in need of response to problems that occur. Thereby he establishes a correlation between positive feedback, path dependencies and institutional change. Positive feedback makes institutional change self-reinforcing, according to Pierson, and these reinforcements can make changes irreversible (pp. 17-53). One of the steps in a sequence of choices can make reforms irreversible if it implies the exclusion of alternative solutions in the next steps. The first step defines the chosen path and excludes other possibilities and thereby makes it difficult to return to previous paths based on the alternatives left out. Pierson (2004) holds positive feedback to be especially useful to explain incremental change in institutions. On the other hand, he develops the concept of institutional resilience: institutions’ capacity and tendency to reject attempts to change and return to original shape. Pierson (2004, pp. 142-153) discusses hindrances to institutional change 1) by coordination problems in institutions, 2) by veto points that are established and intrinsic to the institution and 3) by asset specificity, which connects to the interests of actors involved and that may produce resistance to reform and change. He (p. 150) points to the “institutional matrix” (North, 1990, p. 95) and the “ecology of interconnected rules” (March & Olsen, 1989, p. 170) among several different institutions that relate to and work in a shared context. Such connections between different institutions may be drivers for reinforcement of the stability and status quo in each of them “and make path dependence a common feature of institutional evolution” (Pierson, 2004, p. 150). When positive feedback reinforces change, the institution develops along new paths. This relates to Thelen’s concept of displacement of institutional arrangements. Institutional resilience will on the other hand push the institution back to old paths. Path dependency is the key to understanding whether change or resilience becomes the outcome of a process. Introduction of a new path includes that alternative solutions to the actual problem tend to vanish from the institutional agenda in ways that make it unlikely that those alternatives return on later occasions. Path dependencies will reduce the number of alternative solutions in institutional agendas, and more so over time. Institutions become more and more resilient the longer they exist. Reform and change will tend to become incremental, as the cost of revisions will grow with the strengthening of path dependencies over time (Pierson, 2004, pp. 150-151): “Established institutions thus may create powerful inducements that reinforce their own stability” (p. 150).

4.2 Institutionalization of Working Life The Norwegian Parliament (hereafter the “Stortinget”) passed the first act of labour regulation in 1892. It was the result of worker opposition and strikes in chemical manufacturing. A law enacted on public supervising of the manufacturing industries (“Fabriktilsynsloven”) intended to protect workers from unhealthy working



The First Ideas of Economic Democracy 

 59

conditions, unreasonable working hours and child labour. The preparations and outlines of the law started in 1885 by the establishment of the first Workers Committee (“Arbeiderkommisjonen”) with representatives of employers, ministries and workers’ appointed as members. Due to strong disagreements on rights and concessions to be accorded to workers, the final legal act was a quite limited compromise compared to the proposals from a minority in the committee. This political act illustrates that although labour and capital interests were assumed to be starkly opposed, democratic design of processes was seen as an appropriate tool to come to shared conclusions, even if it took seven years (Falkum, 2008, pp. 176-181). The Workers Committee became a test of ways to reach terms of agreement and introduced the first bricks of the institutional design of working life through dialogues between interest groups. Political authorities intended to pacify a growing labour opposition via a law on labour protection towards exploitation (Bjørnson, 1990; Falkum, 2008). Workers organized their interests in the Labour Union confederation in 1899 (“Arbeidernes Faglige Landsorganisasjon”, AFL, later the Labour Organization, hereafter LO), whereas employers established their confederation of employer associations a year later (“Norsk arbeidsgiverforening”, NAF, later “Næringslivets Hovedorganisasjon”, hereafter NHO). The two organizations were core actors in the institutionalization of working life. They are still the centre of institutional and centralized power and key actors in the design of industrial and labour relations. At the outset of the industrialization of Norway, however, they strongly disagreed on most issues. Despite these conflicts, they managed to develop national agreements in parts of the manufacturing mechanical industry, first in 1902 and an extended version in 1907 that embraced mechanical and metal manufacturing plants. It defined the unions’ rights to negotiate wages, working hours and working conditions (Stokke, 1998, p. 92). This agreement represents a shift in paradigms as the conception of work shifted from a paternalist view of work as a social commitment to the owners to an understanding of work as a purely economic exchange of labour for wages. As these first agreements were broken on both sides by strikes and lock outs shortly after introduction, the state had established a committee to outline rules for handling labour conflicts by 1906. The government presented law propositions on the subject to the Stortinget in 1909, 1912 and 1913 (Bjørnson, 1990, pp. 394-400). In 1915 an act regulating labour conflicts was passed, and the Labour Court and a National Arbitrator were institutionalized in 1916 by formal rules for handling labour conflicts. This is an example of displacement of institutional arrangements (Mahoney & Thelen, 2010).

4.3 The First Ideas of Economic Democracy The Bolshevik revolution in 1917, followed by a German revolution in 1918, fuelled battles over property rights and control of political power all over Europe (Selznick,

60 

 Institutionalization and Dynamic Change of Institutions

1957; Dahrendorf, 1959; Giddens, 1973). In Norway, the first known union demand of a “right to have their share of the management” in private enterprises occurred in the collective wages bargaining in 1918 (Debes, 1919, pp. 198-199). Company owners and the employers’ confederation interpreted this demand from the unions as an attack on their private property rights. The Norwegian Labour Party joined the third Communist International in 1919 but left it in 1921 due to disagreements on the centralization of power and undemocratic leadership of the movement. Neither the LO leadership nor the majority of the LO members were, however, in favour of the Communist International. During the threeyear “fling” with communist ideologies the Labour Party lost about two-thirds of their members. The party turned to social democratic ideologies when they left the international movement in 1921 (Bjørnson, 1990, pp. 544-545). Values, norms and understandings of realities were changing and by no means stable before the Labour Party as well as the LO firmly adopted the ideas of social democracy. The NHO rejected the LO’s demand on managerial participation in private enterprises in 1918. The liberal government guaranteed to support the owners against riots and workers’ attempts to take over enterprises – if necessary with police or military forces. The government and the employers’ confederation supported each other in an alliance and coalition of political and economic power, whereas the Labour Party and the LO represented the opposite alliance. Opposing values, norms and ideologies between the two coalitions dominated the 1920s and led to a series of massive strikes and lockouts. “Economic democracy” was originally a revolutionary idea in the international labour movement. Martin Tranmæl presented the idea of economic democracy at the Labour Party convent in 1918 (Bjørnson, 1990, pp. 525-526). The governmental Worker’s Committee of 1918 (“Arbeiderkommisjonen” of 1918) discussed economic democracy as well as “…the workers’ part in enterprise management”. It was an attempt to design a model that could be acceptable to both sides. The committee transformed economic democracy to a matter of workers’ participation in what they labelled “company councils” and company boards, quite contrary to the Labour Party’s conceptualization. Debes, who was the “workers’ representative” appointed to the committee of 1918 by the government, shows that this idea of economic democracy was discussed in several countries, such as Germany, England, Finland and the U.S., and was supported by, for example, J.D. Rockefeller (Debes, 1919, p. 8). Increased production was a societal and economic responsibility for all participants according to the model (Debes, 1919, pp. 198-199). The model proposed a structure of collaborative and participatory bodies on the industrial and workplace levels. This institutional structure was unacceptable to the LO and the Labour Party, as well as to the NHO. The labour movement still adhered to socialist ideas and viewed “workers part of management” as a moral and just right derived from their contribution to economic wealth. The NHO and the employers insisted on protecting their “managerial prerogative” and private property rights from workers participation and interference.



Designing the Basic Agreement in 1935 

 61

The main institutional designers at the time were employers’ associations, the political institutions and the labour movement. They discussed compromises and enacted rules and laws on how to handle conflicts of interests but did not come to terms on labour participation or union codetermination, as stated above. These rules, however, did not work as intended as long as conflicting values, norms and anticipations guided their acts and behaviour at the workplaces. Conflicting anticipations of realities, interests and social values hindered shared solutions to societal problems. These enduring conflicts harmed national productivity and competitiveness. The situation had worsened by the international economic downturn in 1929.

4.4 Designing the Basic Agreement in 1935 Several attempts to solve the labour conflicts by formal agreements and laws failed during the 1920s. In 1930 the social liberal government established a Labour Peace Committee and invited the NHO and LO to participate. The intention was to make the social partners come to an agreement that could make them cooperate in order to solve differences and avoid further conflicts. The ambition was to improve production, competitiveness and the national economy after decades of destructive labour conflicts. Seim (1972, p. 20) shows that both parties tried to avoid direct representation in the Labour Peace Committee by instead offering some of their individual experts to be advisors to the committee. Maurseth (1987, p. 490) found that the labour union kept their participation secret even from the union congress as late as 1931. The employers’ confederation argued that they would not participate unless the union joined in (Seim, 1972, pp. 22-25). This initiative indicates that the government had given up labour market regulation by legislation and political regulation (Falkum, 2008). The conservative Farmers’ Party (today “Senterpartiet”) entered the governmental offices after their electoral success in 1931. They proposed to regulate the relationship between economic organizations and unions by law and to force them to adapt to governmental policies (Seim, 1972, p. 43). Together with the conservative party, Høyre, they wanted to protect the freedom of employers by forbidding labour strikes and thereby “eradicate the source of contamination of the work life” (ibid.). In contrast, the social liberal party, Venstre, opted for labour peace by collaboration with their social partners. Hence, there were two different strategies among the non-socialist parties. One identified “freedom of the working life” as the freedom of owners and capital by policies to bind “labour”, whereas the other defined labour peace to be necessary to increase productivity and improve national economics. This latter strategy implied tripartite collaboration between employers, unions and the state as the appropriate means to reach labour peace. The conservatives protected private property rights as an institution viewed as more important than the attempts to solve the problems of

62 

 Institutionalization and Dynamic Change of Institutions

working life – in other words, an example of what Pierson (ibid.) called resilience mechanisms that hinder attempts to find new paths to solving societal problems. When the Farmers’ Party government proposed legislation to boycott, block and forbid strikes in “key enterprises” in 1931, the LO left the Labour Peace Committee in protest. The same year the government sent military forces to stop striking workers from preventing scabs to do their jobs at a plant at Menstad. After a strike among construction workers in 1932, the conservative government dissolved the Labour Peace Committee. They lost, however; their proposed “labour hostile” laws boycotting striking workers were rejected by the Stortinget in 1933 (Bjørnson, 1990). That brought the social liberal Venstre back into office. The new government moderated the law proposals after negotiations with the Labour Party, who had supported Venstre in removing the Farmers’ Party government. The social liberal government (Venstre) realized that political regulation of working life by legislation did not lead to the intended results. By inviting employers and unions to negotiate, participate and contribute to labour peace the government shared political power with them. A second consequence of this invitation was that the government had to be neutral to both employers and the union in order to be trustworthy. Hence, the invitation itself broke the old alliance with the employers’ confederation, with the result that the employers changed their interpretation of the political situation and its implications. When the employers and the union confederation chose to formalize rules and relations in an agreement between the two of them instead of adhering to a law decided by political institutions, they strengthened their own political power. Together they grasped the power to set agendas of political importance and to define and decide rules of work life conduct. The political institutions were set aside with less influence on labour policies. The new government proposed a new act to regulate voting over results of bargaining among employers and unions and to extend the authority of the National Arbitrator (“Riksmeklingsmannen”). This proposal split the LO into a social democratic fraction that supported the proposal and a socialist-inspired fraction that strongly opposed it. The social democrats were in the majority, ruled the union congress and won leadership of the confederation. The new LO leaders supported negotiations and parliamentary methods. Both the employers’ confederation and the LO wanted to formalize rules by agreements between them instead of decisions made by political institutions (Maurseth, 1987, p. 495; Seim, 1972). This paved the way for the signing of the Basic Agreement in the spring of 1935. The agreement covered private enterprises with collective agreements on wage formation and working conditions. It consisted of 17 paragraphs on rights to negotiate, how to act under disagreements and differences and union rights to representation and codetermination in the private enterprises that signed it. The agreement stated formal rules and codes of conducts in negotiations. It defined relations between the employers, unions and employees at the national and workplace levels. It defined formal rules for voting in case of disagreement. It contained rules to protect shop



Designing the Basic Agreement in 1935 

 63

floor stewards from unreasonable dismissals. Employers could no longer characterize employees’ behaviour in formal documents of the work relationship when they quit. The Basic Agreement defined principal rules for handling disagreements. The numbers of labour union members had grown considerably since 1920, and so did the votes for the Labour Party in elections. The Basic Agreement boosted the support of unions as well as of the Labour Party. The Labour Party formed their second government in the autumn of 1935 and stayed in office until 1965, except for the five years of German occupation during World War II when they stayed in London. The growing support for the labour movement also influenced the other actors’ anticipations of relations and the distribution of power in politics and economics. Hence, the Basic Agreement implied new roles and relations for all actors in the political and economic systems. By accepting the invitation to contribute to labour peace, and by signing the agreement, the labour movement implicitly accepted the constitutional private property rights and the employers’ administrative prerogative. This was a retreat from the core of their ideological position and source of earlier labour conflicts. The NHO wanted to keep as much influence as possible on the formation of formal rules that regulated and limited their activities. Consequently they chose to make the rules in collaboration with the LO instead of leaving decisions to political institutions. The NHO recognized the LO as an equal partner and accepted the unions’ right to negotiate on enterprise decisions. This was their ideological sacrifice in the compromise between opposing interests. Both the LO and the NHO preferred agreements between the social partners to legislation made by politicians (Falkum, 2008, p. 222). The political authorities now had to handle the social partners with equal respect. The establishment of the Labour Peace Committee was a turning point – the initiative that actually set out new institutionalization processes and an institutional displacement (Mahoney & Thelen, 2010). Participation in the committee changed the understandings of realities in the LO as well as in the NHO. The Basic Agreement confirmed this change in orientations and the will to contribute to a balancing of labour and capital interests. Cathie Jo Martin (this volume) discusses reasons for negotiation myopia: “…cognitive, psychological and strategic impulses that prevent individuals from reaching agreements that could conceivably leave all better off”. One of Martin’s explanations of negotiation myopia is the scope of goals: a focus on distributive benefits makes us blind to ways to enlarge the pool of resources by joint actions (ibid.). Myopia is, in other words, an explanation of institutional resilience as defined by Pierson (2004). Negotiation myopia, as a collective phenomenon where all individuals think and act in the same ways, creates path dependency. Collective actors negotiated the Basic Agreement, however, and the question is how and why myopia was transcended in the case at hand. The Labour Party had slowly moved from ideas of revolution and the concurrence of capitalism to a position still based on some socialist ideas but also the adoption of liberal democratic methods of political selections and allocation of political power in society (Falkum, 2008, p. 284). The

64 

 Institutionalization and Dynamic Change of Institutions

Labour Peace Committee redefined the game. Even though private property rights had been made acceptable, the government had to treat the LO as a participant equal to the NHO, and that conditioned the development of shared understandings and solutions to problems among equalized parties. They could move on from negative to positive feedback mechanisms. Veto points were dropped for new asset specificities that came about through collaboration. Inclusive negotiation put an end to negotiation myopia primarily by new cognitions (Martin, 2013 and this volume). Institutionalization of consensus-oriented industrial relations had begun.

4.5 Institutional Design of Tripartite Structures In the aftermath of the Second World War, the Labour Party, now in power, adopted a programme (“Blåboka”) aimed at redesigning and extending ideas of economic democracy. All the political parties at the Stortinget supported a modified version of this programme (“Fellesprogrammet”). It was simply a plan to restore the country from the damages of World War II (Bergh, 1987; Pryser, 1988). The programme was intended to improve the coordination of economics and politics. Political institutions and democratic arrangements were considered to be what should rule and control the national economy (Bergh, 1987, p. 214). The Labour government established a new structure with a national coordination council (“Samordningsrådet”) on top, with branch and industry councils underneath and production councils at the workplaces. The latter was inspired by American war industrial democratic arrangements (Lie, 1946). The structure was rooted in the ideas of economic democracy from the 1920s and a description of a similar structuring of the interests of working life actors (Debes, 1919; Bergh, 1983, p. 85). Only organized interest groups participated in this new structure. This was, in other words, a corporate channel of influence in parallel with the parliamentary institutions (Falkum, 2008). The national coordination council would handle economic matters before they were taken up by the political parties and the Stortinget. The assumption was that the business and employer organizations in the council would influence the liberal and conservative parties in matters of national economy and thereby give the labour government some control over the opposition in the parliamentary system – an attempt at layering the political institutions (Mahoney & Thelen, 2010). It worked the other way round, as some of the business and employers’ representatives supported the liberal parties’ opposition to the government, in line with the previous alliance between employer interests and the liberal parties. These representatives left the council in 1952. Two years later the Labour government closed down the council and replaced it with a Collaboration Committee (“Samarbeidsnemnda”), which was dissolved in 1987. The first attempt to formalize the corporate structure failed, and the institutionalization of the tripartite collaboration as a model for decision-making took the shape of an informal and closed network. Institutionalization took a different and unintended path.



Changes in Institutional Contexts 

 65

The councils at the branch and industry levels were also tripartite, with participation from the government, employers and employees. Sector councils should work as a decision-making structure advisory to the different industries and enterprises in matters of development, closures, establishments, restructurings and expansion (Bergh, 1983, p. 217). Some of the employers’ associations opposed the councils by repeating their liberal ideas of the necessity of freedom to act, stating that the industry itself and the single enterprises had to control and take care of these matters, not external actors. The basic idea of the sector councils was that cooperation between employers and employees on development would improve productivity, restructuring and economic growth. However, this second formal level of cooperative bodies was not successful either. The actors participated but did not engage with any enthusiasm and ended as an example of institutional conversion (Mahoney & Thelen, 2010). The Production Councils intended to improve productivity in private enterprises by collaboration between management, unions and employees at the workplaces. They were to define goals of production, rationalization and growth. The LO and NHO signed the Production Agreement (“Produksjonsavtalen”) in 1945, formalizing the establishment and activities of the Production Councils. About 1000 such councils existed in 1946, mostly in larger manufacturing plants (Falkum, 2008). From the perspective of institutionalization the actors abandoned the formal designs of a corporate structure. They still supported the ideas of collaboration and dialogues but transformed the institutional structure into something less formal and with looser couplings. The constitution of the structures and the institutionalization took other paths than were intended by the original design.

4.6 Changes in Institutional Contexts The Basic Agreement of 1935 was the end of a more than 30-year period of open labour conflicts and class struggles. In 1946 the Labour government introduced planned economy built on socialist ideas and implemented in national long-term economic programmes, five-year plans and national budgeting. The intention was to regulate economy by politics. The temporary price regulation acts of 1945 (Lex Thagaard) and 1947 (Lex Brofoss) allowed the government to determine how much an enterprise could produce and at what prices to sell. To socialize parts of the manufacturing enterprises was on the agenda as well (Bjørnhaug & Halvorsen, 2009, p. 242). However, exogenous actors and international programmes broadened the economic and political contexts and introduced new windows of opportunities. The Marshall Plan was such an exogenous offer that conditioned the redesign of political programmes and regulations and paved the way for a market economy. The government deregulated restrictions on prices and exports and imports in 1952 in order to join the Marshall programme and to be included in the western economic sphere. The new design was

66 

 Institutionalization and Dynamic Change of Institutions

a mixed economy that combined socialist ideas of a planned economy with liberal ideas of free markets. This influenced the following institutionalization of tripartite structures and eventually led to redesign of the Basic Agreement as well. The change in economic contextual framework was, however, not the only source of institutional change at the time. Ideological differences between communism in the East and liberal democracy in the West was the basis of the Cold War. Whereas the Communist Party in Norway supported the Labour Party government in 1945, they broke out of the alliance and characterized the Labour Party as “…the gatekeeper of imperialism and the distributor of the American fascism…” a few years later (Bergh, 1987, p. 22). The General Secretary of the Labour Party, who had stayed in the U.S. during the war, was highly influenced by the American fear of the growing power of the Soviet Union. He mobilized a campaign against communists as a threat to civilization and society (Lie, 1948). The two parties once more turned on each other in ideological battles. The actors changed values, interpretations of realities, anticipations of each other and eventually the formal rules of the game. The new and international institutional context changed the national institutional context and agendas, broadened perspectives of meaning, gave birth to new interpretations and values and introduced new alternative paths for the national actors with an influence on national institutional design. Bjørnhaug and Halvorsen (2009, pp. 315-342) show that some actors in the labour movement participated in the design and implementation of political surveillance systems and the report of communist activities at large manufacturing plants. The suppression of radical political viewpoints at the workplaces caused workers to fear losing their jobs. The Communist Party lost more than half of their votes between 1945 and 1949. The enduring problem, nevertheless, was the uncertainty of the communist position in the union and at the workplaces. The attempts to discipline the unions and their members at the workplaces had at least three arguments: Firstly, to prevent and reduce opposition and resistance to the labour government’s policies and politics; secondly, to secure centralized control of the labour union; and thirdly, to reduce the communist opposition to collaboration with employers and to implement the cooperation arrangements that the social partners had agreed on in the Basic Agreement in 1935 and the Production Agreement in 1945. This battle between communist versus social democratic ideas lasted throughout the 1950s. Interpretation of realities, what March and Olsen (1995) call meaning, and the cognitive aspects of institutions, clearly affect working life institutions. Domestic as well as international contexts contribute to explanations of the change in cognitive aspects and the development of work life institutions.



Industrial Democracy and Experimental Design of Workers’ Participation  

 67

4.7 Industrial Democracy and Experimental Design of Workers’ Participation When the social democratic design of economic democracy and the formal institutional tripartite structure partly failed and an informal structure with loser couplings emerged. The redesign of a democratic work life started. In line with the Marshall Programme’s demand of a market-oriented economy and the ensuing policies of a mixed economy, a stronger focus on productivity rose. The ideas of economic democracy from the 1920s and 1940s were transformed into ideas labelled “industrial democracy” in the late 1950s. Arguments about just distribution of representation and influence between employers and employees dominated the first, while productivity improvement was the strategic intention of the latter (Falkum, 2008). In 1961 the LO and NHO established a committee (Felleskomiteen LO/NAF) to analyze the possibilities for and potential impacts of workers’ representation on company boards, headed by professor Einar Thorsrud. By 1953 the Labour Party had called for legislation giving workers the right to representation on the boards of enterprises but did not get support in the Stortinget. The social partners anticipated this law proposal as a governmental attempt to retrieve the power in labour politics that they had achieved by the Labour Peace Committee in 1930 and the Basic Agreement in 1935. The establishment of this committee, and its analysis, was the social partners’ response to the law proposal they did not support. Thorsrud’s conclusion was that workers did not carry the necessary knowledge to be members of company boards (Thorsrud & Emery, 1964). Instead of employee representation in company boards, the committee proposed “broad participation”, which referred to every individual employee’s participation in and responsibility for enterprise development. Inspired by Eric Trist and his analyses of the British mining industries at the Tavistock Institute, experiments with self-managed groups of workers were set out in five large Norwegian manufacturing plants. This turn from union codetermination at the company level to employee participation in organizational development represented a new definition of “industrial democracy”. Unions and shop floor stewards were involved in the design of the experiments in each enterprise (Thorsrud & Emery, 1970). “Partly self-managed groups”, however, was more of a management concept than a model for workplace democracy (Falkum, 2008, p. 120). Thorsrud’s contribution was first and foremost his introduction of Human Relations theories to Norwegian work life, albeit 30 years after Elton Mayo’s famous Hawthorne experiments (Falkum, 2008). At the beginning of the 1960s a new battle on distribution of power in labour politics rose between political and work life actors. The government and the Labour Party wanted political control and proposed to anchor the rules and regulations in the political institutions but lacked support from all the other political parties in the Stortinget. A problem for the government, however, was that the leadership in the LO, in line with the employers’ confederation, wanted to constitute the working life rules

68 

 Institutionalization and Dynamic Change of Institutions

and regulations in private sector by agreements, not through legal acts, even though the labour unions congress in 1965 supported the introduction of board representation in the public sector by legislation. At that time, the labour union and the employers’ confederation had already started negotiations over a reform of the Basic Agreement. In a speech to the Iron and Metalworkers’ Union in 1966 Tor Aspengren, Secretary General of the LO, stated: We can never make a viable work life democracy just by making laws or by including new paragraphs in the Basic Agreement. It can only grow gradually in a positive climate in the companies, as part of a natural lifestyle. Therefore, our enlightening, educative and work environmental policies are crucial in our work. (Aspengren, quoted in Thon, 1966, p. 85, my translation)

The quote presents an almost sociological analysis of institutionalization. The head of the labour union did not give strong support to the idea of collaboration with management only at the company level; he forcefully underscored that the practice of codes of conduct is more important than the rules that constitute these codes. However, a constitution of the rules by law would have to be universal and embrace all employees, not only union members. In the Basic Agreement, in contrast, the labour union elects the representatives to democratic bodies at the workplaces among their own members. In other words, the constitution of labour rights by laws would reduce the power of the labour union and weaken its position towards employers in negotiations and bargaining at the workplaces as well as nationally. The political intention to regulate work life by legislation would imply a major change in the institutions. In this case, the proposed change was hindered by institutional resilience. The core value of the employers was the freedom to decide and act in favour of their own (economic) interests. This was in line with the ideologies of the liberal parties at the Stortinget in formal opposition to the Labour government. The ambition of the government was to regain the power to rule labour politics that had been gained by employers and unions by the Basic Agreement as regulatory rules. The labour union would avoid power losses at the workplaces that would follow universalistic legal principles. The government wanted to regulate work life by legislation while the work life actors wanted to do so by less binding agreements. This was the institutional change at hand. An extra boost to the debate over employee board representation, as well as to the debate over rules by legislation versus agreements, came with the major accident involving 23 casualties in a coalmine in Kings Bay at Svalbard in 1962. The Labour Party argued that employee representation could have given the board of the mining company information that could have prevented the catastrophe. The liberal opposition now supported the implementation of employee board representation in state-owned enterprises. In 1964 and 1965 the social partners expected the collaboration experiments to soon be concluded, long before the formal deadline, and that these would support the governmental intentions of regulations by law (Bergh,



The Change of the Basic Agreement in 1966 

 69

2009). These events resulted in a hastening of the revision of the Basic Agreement. The employers and the union finished a new version very quickly in the hope of avoiding regulations by laws made by the political institutions.

4.8 The Change of the Basic Agreement in 1966 Neither the employers’ confederation nor the labour union wanted to have employee representation on company boards. The employers anticipated the governmental proposal as another attempt of political interference in private property rights and their managerial prerogative. However, the employer prerogative was never supported by any formally decided rule, but custom and common acceptance legitimized it. Employers feared that employee representation on boards would reduce the owners’ freedom to act. The union, however, was against employee representation due to fear of co-optation. They would be a minority with one-third of the board members and reasoned that the employee representatives would easily become hostages for decisions that they opposed but were unable to stop. Both the union and the employers’ confederation referred to Thorsrud and Emery’s (1964) analysis and argued that employees would lack competence to do a proper job as board members in private companies. The working life partners had to come up with an alternative to a legal act on employee board representation in order to avoid it. The Basic Agreement of 1966 defined a Company Assembly (“Bedriftsforsamling”) as an alternative. Unions, employees and managers would discuss issues of business and organizational matters. The assembly was conceived as advisory to the company boards (Hagen, 2010). This was an institutional change but of less significance than employee representation on company boards. In 1966 the social partners extended the Basic Agreement and divided it into a part A and a part B. Part A presents the unions’ rights to collective agreements, and co-determination in business and enterprise matters. The union representatives/ shop floor stewards represented all employees in the collaboration with enterprise management. This was the basic democratic arrangement at the workplaces. Revisions of the agreements during the 1950s, and especially in 1962, combined these rules in a more precise formulation (Article 9). The 1966 revision formally justified the democratic arrangements with the intention to increase productivity. The first agreement of 1935, on the other hand, had justified working life democracy according to the need for labour peace, collaboration and human relations, despite the fact that economic crisis, low productivity and national competitiveness was the governmental outset for the labour peace committee in 1930. Union representatives still hold part A, and especially Article 9 about union co-determination, as the crucial rule by which to legitimize workplace democracy. The partners agreed on extensions of part A in 1966 but did not introduce new

70 

 Institutionalization and Dynamic Change of Institutions

principles regarding acting or codes of conduct compared to the 1935 version – only more detailed descriptions of procedures for collective bargaining and collective agreements. Part B, however, was completely new and focused on the participation of the individual employee in organizational development. Part B defined the individual employee’s rights to participate in the development of their own work situation and influence the distribution of tasks, work performance and relations at work and also included commitments to participate in the improvement of productivity. Companies with 50 employees or more were obliged to establish work councils (“Bedriftsutvalg”), which constituted a revision, renaming and extended version of the Production Councils from 1946 (in part B of the Basic Agreement). Union representatives and managers were entitled to participate in these councils to discuss organizational changes. Part B stated this principle. While part A of the agreement defined the unions’ rights to negotiate and bargain, part B defined the rights to co-determination and collaboration on the implementation of enterprise changes and development. A two-stringed democracy at the workplaces developed: one based on negotiations, rights and duties and one based on rights to collaboration in organizational development. The first was collectively oriented, while the latter focused on individual participation. The social partners picked up most of part B from the productivity agreement in 1945. The commitment to contribute to increased productivity obtained stronger legitimacy in the Basic Agreement than in the previous sub-agreements. The rules of the institutional collaboration changed. Part B formed the base for later development of the democratic arrangements at the workplaces (Falkum, 2008). The application of democratic concepts like participation, co-determination, cooperation and influence are vague and ambiguous in the 1966 version of the Basic Agreement. The concepts have no strict definitions, and they are often substituted for each other in the formal text as well as in practice. Thus, we find various interpretations and intentions in the implementation of the rules (Colman et al., 2011)

4.9 Positive and Negative Feedback of Institutional Change Positive feedback explains why institutions change and how institutional change can be irreversible (Pierson, 2004). Creating, formulating, agreeing on and deciding rules has not been a great problem in the institutionalization of the working life. The first agreements in 1902 and 1907 evolved through the shared efforts of the institutional designers: the LO and NHO. The actors did not follow these rules in practice, however, and ensuing labour conflicts gave negative feedback to the rules and intentions they had agreed upon. Mutual negative feedback and sanctioning aborted, hindered and stopped the institutionalization processes until 1935. The employers, the union and the political institutions reached some agreements like the establishment of the National Arbitrator and the Labour Court despite the conflicts of the capitalist and



Institutional Design and Designers 

 71

working classes in the first decades of the 1900s. The Basic Agreement was a set of rules that implied positive feedback to consensus and negative feedback to conflicts. This turned the trend of labour conflicts. Positive feedback not only correlates with the successful change of institutions; it is also a major condition for institutionalization as such. Negative feedback can explain how attempts to change the institutional design, rules and structures fail. The institutional rules prescribe and predict acts and behaviour in social systems according to institutional theory. We have to add to Pierson’s concepts the fact that social systems also use institutional rules to evaluate and sanction institutional acts and behaviour and attempts to change established values, norms, rules and conceptions of situations, structures and relations – that is, the way the institutions work. How institutions change and develop can be explained by institutional evaluation and sanctioning mechanisms. Illustrations can be found in the governmental design of the corporate institutional structure of the 1940s and 1950s. It ended in an informal and less binding structure after negative feedback and withdrawal from employers and their associations. Large parts of the intended structure terminated after a decade or so. The institutions bounced back to their original shape after quite serious attempts to change them. Negative feedback mechanisms and sanctioning are implicitly important variables to explain such institutional resilience. The governmental establishment of the Labour Peace Committee in 1930 was intended to cope with the negative feedback and veto points to labour politics and the distribution of wealth from the labour conflicts of the 1920s. When both unions and employers’ confederations joined the committee, it gave a positive feedback that in the end resulted in the Basic Agreement as one of the most important institutionalizations of the working life.

4.10 Institutional Design and Designers Institutional resilience becomes stronger the older the institutions are, according to Pierson (2004). That implies that the institutional design becomes more widely accepted and influences more of the acts and behaviour in social systems as they grow in size and age. Pierson (ibid.) points to this to explain why change is less likely to occur in institutionalized than in goal-seeking, market-oriented systems and organizations. When the labour union and the employers’ confederation originally designed the Basic Agreement, the government had initiated the process but had little to do with the design and the later redesign. On the contrary, the political institutions tried to recover power in labour politics on several occasions. The basic principles in the first agreements was to set procedures for the development of shared understandings of situations, reaching consensus and handling disagreements, differences and conflicts. This is still the basic principle. Originally, the Basic Agreement embraced only the manufacturing industries. New unions and employers’ associations have emerged and they have established several new basic agreements

72 

 Institutionalization and Dynamic Change of Institutions

in the public sector and service industries. These later agreements build on the Basic Agreement in private manufacturing industries. Consequently, the principles of the Basic Agreement cover the majority of employees in the labour market today. The original rules have been changed, extended and modernized over the years, but the basic principles like the rights and obligations of the actors in this major social system are, basically, the same. We have to distinguish between the initial design of an institution and the later development and redesign of the same institution. It is easier to observe who is behind the design in the initial phase than in the later development of institutions, especially in old institutions in large social systems. Geopolitical changes as well as changes in domestic political and economic organizations influenced the institutionalization of the working life from 1935 to 1965. These contextual changes did not result in a special redesign of the rules in the Basic Agreement. It was the understandings of situations, political and economic contexts and interpretations of realities and activities – that is, the institutional “meaning” – that changed first. The new cognitive anticipations of the contextual framework and eventually the institutional redesign led to a new institutional structure based on more informal and less binding commitments and relations. The corporate and tripartite relations designed by the Labour government and supported by a majority in the parliamentary system in the late 1940s were negotiation-oriented and hierarchical. The redesigned structure that grew out of the contextual reframing and negative feedback from participants was collaborative and consensus-oriented. This is the structure that Rokkan (1966) labelled “corporate pluralism”. The redesign of the institutional structure was the result of trials and errors, negative and positive feedback from the different participants and a focus on what would work in practice rather than formal rules and strategies. The sequences in the described change process are: 1. Class struggles and negative feedback from 1900 to 1935 2. Labour peace by the Basic Agreement in 1935 3. Establishment of a formal hierarchical institutional structure 4. Contextual reframing (the Cold War, Marshall Plan) 5. Development of new and shared understandings of realities (new cognitive structure) 6. Redesign and implementation of the institutional structure 7. New institutional acting, behaviour and practice 8. Rivalry on constitutional methods (laws versus agreements) 9. Revision of rules in the Basic Agreement in 1966 The change of institutions is not necessarily a matter of constitutions and formal rules. Institutions can change without a change in formal rules (Mahoney & Thelen, 2010). It depends on the contextual reframing, the development of a shared cognitive structure (understanding of realities), the exchange of feedback and a redesign of procedures and relations between the actors. Old rules legitimate new ways to act



Path Dependencies, Resilience and Institutional Matrix 

 73

and respond – that is, rules are interpreted and applied in new ways without changes to the rules themselves but more by the ageing of institutions. This explains why institutional change can be very hard to observe as it takes new paths tacitly and without formal decisions.

4.11 Path Dependencies, Resilience and Institutional Matrix Path dependencies are patterns that grow and become principles for ways to anticipate situations and respond to them, almost automatically. The choice of a new path can be the outset of institutional change, but the system can also drop it in favour of existing procedures and routines. Path dependencies make systems stable and predictable, for better or worse. That is a question of the qualities of institutional redesign. In this perspective, institutional resilience becomes a mechanism to test out new paths and new solutions to occurring problems and affects institutional change. Resilience implies the evaluation of a new institutional design and negative or positive response to it. Institutional resilience mechanisms may result in 1. Positive feedback to new paths in the system and support for institutional change 2. Ignorance of new paths, change introduced without support in the system and the institution bouncing back to established paths 3. Negative feedback to new paths and the system’s rejection of a new design or chosen paths Institutional resilience tests out proposed changes, supports acceptable new designs and chosen paths and modifies and redesigns proposals to become acceptable in the system. Institutional resilience results in the rejection of and resistance to institutional change but can also contribute to dynamic change of institutions. The development of working life institutions was set out in a time and context of a less institutionalized society. Some institutions were in place, however. The battles between capital and labour were highly influenced by the private property rights formalized in the Constitution and practiced long before that. The urge of the labour movement to reduce private property rights was a main element in the class conflicts throughout the beginning of the last century. Private property rights, however, was a strong institution which the labour movement accepted as a legal principle by signing the Basic Agreement in 1935. The actors formed codes of conduct that channelled differences into appropriate ways to overcome negotiation myopia and to reach consensus (Martin on myopia, Engelstad on private property rights, this volume). To take a broad perspective: The Constitution of 1814 designed the political institutions of the Norwegian state. The introduction of parliamentary governance in 1884 redesigned the institutions by choosing new procedural and structural paths for political decisions. The government as well as the Stortinget interfered in the formation and design of work life institutions all the way through the institutionalization

74 

 Institutionalization and Dynamic Change of Institutions

processes, and they still do. The history of the Basic Agreement reveals relations and institutional structures that link economics to politics as two mutually dependent systems. This became obvious in the development of the tripartite collaboration and the corporate pluralism that link economic and political interests. The matrix of institutions is an important part of the context of any process of institutionalization. The principles of the Basic Agreement had important effects on other aspects of working life and social policies. After 1972 several laws have regulated employee board representation, the work environment and working conditions. Welfare state arrangements like the public universalistic education systems, health care services and social insurance systems developed through the 1950s and later on in seemingly continuous revisions and reforms adapting to contextual changes. The Basic Agreement of 1935 rose through the labour movement’s opposition to economic and political elites protected by the political constitution and institutions. The redesign through the 1950s and 1960s allowed new institutions to evolve with the tripartite collaboration as the arena for achieving consensus on economic and political issues. The institutional matrix has grown significantly. This chapter has focused on the opposing interests and powers of labour and capital and the institutionalization that balances them out by rules and codes of conduct. The liberal political parties and the employers’ association’s confederation protected their interests regarding the rising working class through the early 1900s. The labour movement intended to redistribute wealth in society by attacking private property rights as well as political institutions. None of the actors had the power to dominate and set the agendas in this evolving system. Institutionalization appears to be a more functional and appropriate way to set the agenda, solve problems and choose the paths in systems where no one has enough power to do so alone or in coalition with other actors. Institution means distributed power to design and implement new paths in society.

References 

 75

References Bergh, T. (Ed.) (1983). Deltakerdemokratiet. Demokrati og samfunnsstyring. Oslo: Universitetsforlaget. Bergh, T. (1987). Storhetstid (1945–1965). Arbeiderbevegelsens historie i Norge. Oslo: Tiden Norsk Forlag A/S. Bergh, T. (2009). Kollektiv fornuft. LOs historie. Volume 3, 1969–2009. Oslo: Pax. Bergh, T. (Ed.) (2010). Avtalt spill. Hovedavtalen LO- NHO 75 år. Oslo: LO-NHO. Bjørnhaug, I., & Halvorsen, T. (2009). Medlemsmakt og samfunnsansvar. LOs historie, volume 2, 1935–1969. Oslo: Pax. Bjørnson, Ø. (1990). På klassekampens grunn (1900–1920). Arbeiderbevegelsens historie i Norge. Oslo: Tiden Norsk Forlag A/S. Colman, H. L., I. Stensaker & J. Tharaldsen (Eds.) (2011) A merger of equals? The integration of Statoil and Hydro’s oil and gas activities. Bergen: Fagbokforlaget. Dahrendorf, R. (1959). Classes and class conflicts in industrial societies. Stanford: Stanford University Press. Debes, I. (1919). Økonomisk demokrati. Arbeidernes andel i bedriftsledelsen. Kristiania: Steenske Forlag. Dølvik, J. E., T. Fløtten, G. Hernes, & J. M. Hippe (Eds.) (2007). Hamskifte. Den norske modellen i endring. Oslo: Gyldendal. Falkum, E. (2008). Makt og opposisjon i norsk arbeidsliv. Dissertation for the dr.philos. degree. Oslo: Universitetet i Oslo. Giddens, A. (1973). The class structure of the advanced society. New York: Harper & Row. Hagen, I. M. (2010). Det mektige mindretallet. Ansatterepresentasjon i styret mellom Corporate Governance og Industrial Relations. Dissertation for the PhD degree.Oslo: Universitetet i Oslo. Hall, P. A. (2010). Historical institutionalism in rationalist and sociological perspective. In J. Mahoney & K. Thelen (Eds.) Explaining institutional change. Ambiguity, Agency, and Power (pp. 204-224) Cambridge: Cambridge University Press. Lie, H. (1946). Produksjonsutvalgene i de forente stater under krigen. Oslo: Arbeidernes opplysningsforbund. Lie, H. (1948). Arbeiderbevegelsen I de forente stater. Oslo: Arbeidernes opplysningsforbund. Mahoney, J., & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change. Ambiguity, Agency, and Power (pp. 1-38) Cambridge: Cambridge University Press. Mahoney, J., & Thelen, K. (Eds.) (2010). Explaining institutional change. Ambiguity, Agency, and Power . Cambridge: Cambridge University Press. March, J., & Olsen, J. P. (1989). Rediscovering institutions. The organizational basis of politics, New York: The Free Press. March, J., & Olsen, J. P. (1995). Democratic governance. New York: The Free Press. Martin, C. J. (2013). Conditions for successful negotiation: Lessons from Europe. In J. Mansbridge & C. J. Martin et al. (Eds.), Negotiating agreements in politics. (pp. 121-144). American Political Science Association. Task Force Report. Maurseth, P. (1987). Gjennom kriser til makt (1920–1935). Arbeiderbevegelsens historie i Norge. Oslo: Tiden Norsk Forlag. Moene, K. (2007). Den nordiske modellen. In: Tariffavtalen som formet Norge. Verkstedoverenskomsten 100 år. Oslo: Pax. North, D. C. (1990). Institutions. Institutional change and economic performance. Cambridge: Cambridge University Press. Pierson, P. (2004). Politics in time. History, institutions, and social analysis. Princeton: Princeton University Press.

76 

 References

Pryser, T. (1988). Klassen og nasjonen (1935–1946). Arbeiderbevegelsens historie i Norge. Oslo: Tiden Norsk Forlag. Rokkan, S. (1966). Norway: Numerical democracy and corporate pluralism. In R. A. Dahl (Ed.), Political oppositions in western democracies, New Haven: Yale University Press. Seim, J. (1972). Hvordan Hovedavtalen av 1935 ble til. Oslo: Tiden Norsk Forlag. Sejersted, F. (2011). The age of social democracy: Norway and Sweden in the twentieth century. Princeton: Princeton University Press. Selznick, P. S. (1957). Leadership in administration. A sociological interpretation. Berkeley: University of California Press. Stokke, T. A. (1998). Lønnsforhandlinger og konfliktløsning, Norge i et skandinavisk perspektiv. Dr.polit. dissertation, Fafo-rapport 246. Oslo: Fafo. Thon, S. (1966). Samarbeide, medbestemmelsesrett, bedriftsdemokrati. Oslo: Elingaard Forlag. Thorsrud, E., & Emery, F. (1964). Industrielt demokrati. Representasjon på styreplan i bedrifter? Oslo: Universitetsforlaget. Thorsrud, E., & Emery, F. (1970). Mot en ny bedriftsorganisasjon. Eksperimenter i industrielt demokrati. Oslo: Tanum.

Inger Marie Hagen

5 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail In Norway, as in a number of European countries, employees are entitled to elect representatives to the company board. In addition – and in contrast to other countries – a Company Assembly (“Bedriftsforsamling”) is prescribed by law in companies with more than 200 employees. One-third of this assembly is elected by and among the employees. The assembly was an important part of the debate prior to the co-determination reforms in 1972 and the LO1 was a strong advocate for the arrangement. Nevertheless, hardly any assemblies are left today, whereas board representation is well established and uncontroversial. The establishment of institutions as well as institutional change are important keywords when the “rise and fall” of the Company Assembly is examined; why did one part of the industrial democracy reform fail and the other prevail? Participation schemes have contributed to the Nordic success – or, maybe more precisely, they are important parts of the successful institutional framework. The different arrangements of collaboration are aimed at different areas (wages, health and safety, decision- making processes, etc.) but they all have the twofold aim of increased production and efficiency along with the democratization of working life (Engelstad et al., 2003; NOU, 2010:1). The conflict between labour and capital is not defeated or won by one of the parties but curbed through different arrangements and “rules of the game” (Falkum, 2008). Working life collaboration may be seen as an institution in its own right, in line with Scott’s (2008) framework: a legal pillar (labour and company law and collective agreements), a normative pillar (norms on working life behaviour) and a cognitive/ cultural pillar (a certain perception of fair working conditions and division of power and social conventions at work). The company duty of establishing a Company Assembly and the employees’ right to demand representation on the board (BLER2) represent the last major reform3 in the struggle to regulate the other fundamental institution in the labour market: property rights.

1 The Norwegian Confederation of Trade Unions is the oldest (1899) and largest confederation in Norway (900,000 members in 22 different unions). 2 BLER (board level employee representatives) is the most common abbreviation in English. BLER is used when I refer to the arrangement as such, while employee reps are used when referring to the representatives. 3 The Working Environment Act (WEA) in 1977 was important, but the area as such – working environment – has been subject to legal regulation since 1892. The 1977 act increased the importance of social partners and introduced important individual rights to participation, but the legislation did not introduce new actors into the system. The 1972 reform incorporates the owners into the collaboration model. © 2015 Inger Marie Hagen This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

78 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

The Company Assembly is a part of the governance structure of the company – below the General Assembly and above the board of directors. The competence of the assembly covers some issues otherwise in the hands of the General Assembly as well as of the board. Thus, both the assembly and BLER constitute limitations to shareholder democracy (or shareholder rights). The assembly reduces the importance of the General Assembly, and moreover, shareholders may only elect two-thirds of the board.4 Both arrangements were introduced in 1972 after decades of heated debate. Today only a handful of assemblies are left, partly due to the legal possibility of dissolution introduced in 1988. In contrast, BLER – part of the same debate and subject to the same opponents and ideological resistance – is well established and uncontroversial among the employees, the employers and representatives of large shareholders (Hagen, 2014). In the wording of a member of the board in a large company: All serious shareholders and managers are pro board representation by the employees. They look upon the arrangement as a natural part of the system. (my emphasis, see section 4 for methodology)

If BLER found their way into the system, the Company Assembly remains as an odd and rare part of the governance structure. Several questions arise: Did the reform in 1972 represent any institutional changes in the labour market model? Did it affect the collaboration between the social partners at different levels? Both the question of why the two elements differ in their institutionalization process and, if so, how institutional changes occurred, are the main topic of this chapter. Employee representation at the shareholder level may be perceived as having two different starting points: i) the arrangement was the last piece in the jigsaw puzzle of employer–employee collaboration or ii) the reform opened up a whole new field of regulations; by employee participation in company matters the dream of societal control of economic life could be fulfilled in a modern non-revolutionary way. The jigsaw puzzle perspective suggests looking at the reform as an example of layering – new rules in order to increase employee participation were added onto the regulation of the relationship between capital and labour. The second starting point has several implications, and we need to ask whether major changes in the collaboration have taken place: Has the power balance shifted on any level? Has the willingness of the social partners to search for compromises been affected? And further: Is it possible to trace any effect at the societal level and have the arrangements succeeded in any institutionalization of regulatory force?

4 Thirty employees are needed in order to demand one employee rep. If the number of employees exceeds 50, the employees may demand one-third of the board. The same thresholds apply to groups (boards of parent companies).



Spheres of Democracy  

 79

Moving to more actor-oriented approaches and as emphasized by several historians (see Bergh, 1983), the debate on company assemblies and representation on the board took place in a small circle of top trade unionists, researchers and politicians who succeeded in implementing the legal reform; thus, was it an example of exogenous shock to the other actors whose positions were anchored in the collective agreements. After these introductory remarks I present a brief review of the Norwegian collaboration system. In the second section I focus on the arrangements found in Norwegian company law and how they fit into the larger picture. The third section presents some empirical findings on actual coverage (how many companies comply with regulation) and perceptions of the arrangements. The fourth section focuses on different understandings of the outcome by returning to the features and logic of the institutional framework. A few other “failed” arrangements are brought into the analysis: the 1920 collaboration committees, the post World War II Production Committees and the attempt in 1980 to introduce parity both in the Company Assembly and on the board (the Skytøen Committee). In the concluding section I elaborate on the two main arguments: Institutional arrangements prevail when they are in line with the overall logic of the system, or in other words, the element needs to fit into the larger institutional framework. As both the Company Assembly and BLER in several respects are “alien” to the regulatory logic, additional understanding is needed to make sense of their different fate. Mahoney and Thelen (2010) emphasize the importance of discretion in both the interpretation and enforcement of rules as possible sources of change: “Conversion normally occurs when rules are ambiguous enough to permit different (often starkly contrasting) interpretations” (ibid., 2010, p. 21). My point of departure is somewhat different; I argue that conflicting interests may maintain institutional elements and thereby prevent conversion. When conflicting interests each (or independently) value the way the institution limits and renders action possible, the lack of compatibility with the overall logic of the institutionalized collaboration system is less important. Or (almost) in the famous words of Selznick (1957/1997): An arrangement may remain stable when it is infused with conflicting values or goals. I hypothesise that the BLER arrangement was able to survive because it is valued by both parties, albeit based on conflicting interests.

5.1 Spheres of Democracy The path from the first collective agreements in the early 20th century to the extensive framework of contemporary Norway may also be conceived as a history of property rights regulation – by law or collective agreements, at the central or company level and the distinction between direct and indirect regulation. The historical conflict between labour and capital in Norway has two stages: i) revolutionary in order to

80 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

eliminate property rights and ii) reformist in order to limit the privileges dedicated to property. Even if key features were in place5 the Basic Agreement from 1935 is often used to date the transition from conflict to collaboration. Nationwide collective agreements ensure worker power vis-à-vis local capitalists but do not prevent, for example, increased demand for dividend to shareholders instead of investments in the company. If the Basic Agreement and the first Labour Party government6 put an end to the revolutionary period of the labour movement, the overall aim of a socialist (or social democratic) society did not disappear. A major programme on socialization was presented by the Labour Party in 1945, proposing the establishment of collaboration committees at the central, industry and local levels (on the Production Committees, see below) in a modernized corporatist structure. Until the early 1950s the Labour Party government established and defended extended regulation on the import and export of goods and capital (“rasjoneringslovene”). Most historians agree that these attempts failed (Sejersted, 2011). Alternative ways of achieving social democratic limitations on property rights had to be established. Heiret (2003) presents the two-level model that emerged during the 1950s and 1960s, distinguishing between economic democracy and industrial democracy. Economic democracy refers to the national political system and thus is subject to political responsibility, whereas industrial democracy refers to the division of power at the company level. As part of economic democracy the capital side is regulated at the national level and in this the national trade unions have an important role. The trade unions trust the state to curb private property: “By the help of the state the conflict of interests should be controlled so that both parties in the labour market contribute to a large-scale modernization of Norwegian industries and thus provide wealth and welfare for all” (Heiret, 2003, p. 114, my translation). The main instruments of regulation were indirect, such as taxation, licenses on shareholding, capital regulations and interest rates. The basis for participation in economic democracy is citizenship and membership in trade unions and trade union influence at the political level. Wage negotiations are an important part of economic democracy. Centralized (or sector-wise) negotiations limit one of the most important consequences of shareholding – the right to dividend. Powerful trade unions “extract” the profit sharing issue from the local level; the pay rate is set at the national level. And thus, the main issue in the conflict between labour and capital – how to distribute surplus – is placed at the national level and in the hands of the social partners, not of (individual) shareholders.

5 Whereof the first national collective agreement in the metal industry from 1907 and the Act on Labour Disputes from 1915 are key events. 6 Or to be precise, the second. The Hornsrud government from 1928 was however dismissed on the inaugural address after only 18 days in office.



Company Assembly and Employee Representation on the Board 

 81

In contrast, industrial democracy is located at the local level; managers and the managerial prerogative, not the owners, are the opponents of the trade unions. The rules of the game are based in collective agreements. The right of the trade unions’ representatives to be informed and consulted on both the day-to-day activities of the management and any major changes in production is found in Article 9 in the Basic Agreement and might be conceived as the core of the collaboration. Locally, the parties collaborate within the limits set by the managerial prerogative. Historically, owners, or shareholders, were perceived as capital providers with little or no part in the decision-making process. Decisions were made either by management or in the social dialogue. The prominent historian Jens Arup Seip put it this way (in the words of Sejersted [2001, p. 54]): When Jens Arup Seip stated that the CEOs – to their amazement – could argue that they were in heaven, this applied to management only and not to capitalists. The bureaucratic managerial capitalism managed to resolve the conflict of labour and capital in a way that was accepted by most people and thus contributed to lessen the level of antagonism in Norway. (my translation)

The model, the role of the social partners at the central and local levels, is an important backdrop to the arrangements in question.

5.2 Company Assembly and Employee Representation on the Board Both arrangements are anchored in the Limited Liabilities Companies Acts (hereafter Companies Act7). The assembly is, as already mentioned, a company body between the General Assembly and the board of directors. Two-thirds of the members are elected by the General Assembly and one-third by the employees. The main task is to elect the members of the board. However, and by rather complicated legal provisions, it is ensured that one-third of the board members are elected by and among the employees of the company. In addition, the Company Assembly is obliged to supervise the CEO and the board and – after the proposals from the board – to make decisions involving “a substantial size of the company resources” and “changes and reorganisations with impact on the number of workers” (ASAL §6-38 [4]8). In 1988 changes in the Companies Act made it possible – if a majority of the employees or trade unions with at least two-thirds coverage agreed – to dissolve the assembly. The tasks and

7 The paragraphs in question cover the limited companies as well, thus the Companies Act. 8 ASAL= Act on Public Limited Companies. Over the years a 10 per cent rule of the thumb has been established, for example, if the changes concerns more than 10 per cent of the work force or 10 per cent of the company capital, approval of the Company Assembly is needed. See Lindkjølen (2010) for a legal review of the tasks of the assembly.

82 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

obligations are then transferred to the General Assembly (electing board members, supervising the board) or to the board (supervising management and making the decisions). As compensation the employees may extend their representation on the board by electing one additional board member (e.g., one-third of board members plus one) or appointing two observers. The history of the Company Assembly is closely entangled with the history of board representation. In the first Labour Committee in 1885 worker representation in company management was one of the issues (Falkum, 2008).9 Board representation more specifically was introduced in the “party programme” of the Labour Party in 1945 but was then sacrificed in order to establish a common “Common Programme” (Fellesprogrammet) for post-war reconstruction, supported by all political parties. The issue appeared anew in the Labour Party programme in 1954 but then disappeared until the early 1960s. The history of the assembly as such starts in 1961 when the Aspengren Committee10 was given the mandate to report on a prospective “Democratic Company Assembly”. In 1963 the committee recommended creation of a Company Assembly but refrained from demanding board representation. In major parts of the trade unions board representation was met with scepticism, based on both ideological arguments and fear that the representatives on the board would end up as a hostages (Grønlie, 1977). The committee maintained that legislated board representation would “weaken the idea of the Company Assembly as a central democratic body; a body given the right to elect board members and with the overall responsibility [for the company]” (Jacobsen, 1977, p. 15, my translation). This is important because it accords the Company Assembly a central part in the governance structure, a concern often raised by company actors today. The path to representation in company bodies necessarily took time. Both the employers’ association (NAF11, now NHO) and the conservative parties were fierce opponents of any such arrangements. Property rights arguments were important. In the end (i.e., at the end of the 1960s) four different views were present: i) The Labour Party – wanting legally based board representation, ii) LO – sceptical towards representation on the board and the only advocate of the Company Assembly, iii) The

9 Another committee was set up in 1918. Bergh (1983, p. 85) emphasizes the ‘’exceptional thoroughness” given to the issue and claims that a substantial part of the later debate has mainly been a repletion of the arguments from 1918. 10 A joint LO and Labour Party committee. Tor Aspengren was the head of the Metal Workers Union from 1958 to 1966 and the head of LO from 1967 to 1977. He is often named ”the father of industrial democracy in Norway”. 11 Norsk Arbeidsgiverforening, later Næringslivets Hovedorganisasjon – the oldest (1900) and largest employers’ confederation in the private sector.



Company Assembly and Employee Representation on the Board 

 83

employers – opposed to any reform in the area and iv) The conservative political parties who were “captured” by the Kings Bay incident12 to support worker representation. Bergh (1983, p. 115) argues that the “unknown construction” of a Company Assembly was more frightening to the employers’ side, and thus they ended up with supporting board representation “as a necessary concession, given the particular situation” (ibid., my translation). It was also important that the right to demand and elect the representatives was based “by and among all employees’ and not given to the trade unions. Sejersted (1984) maintains that the fear of socialization was important to the conservative parties. In the end representatives elected by employees was a far better arrangement than directors appointed by the Government. For LO the Company Assembly was meant to be a transitional arrangement on the path to board representation (Bergh, 2009) – in other words, to be a learning arena. The debate in Parliament (Stortinget) ended by both the Company Assembly and the BLER being passed. Bergh (ibid.) also emphasizes the “political game” in Parliament as an important factor for understanding why both arrangements were introduced. In 1976 the same provisions were introduced at group level – for example, the employees might demand representation if the total number of employees (parent plus subsidiaries) meets the same thresholds that apply to individual companies. To summarize – the arrangements deviate from the logic of the collaboration system in several ways. First, no social partners are involved; the arrangement has a legal foundation in company law. Employee representatives are elected by and among the employees. This has several consequences; all employees, not only members of the trade unions, take part and have a say. The representatives do not represent the interests of the local, or the national, trade union, and thus the strength of the local union is irrelevant.13 Further, local trade unions may call upon support from the national level; the employee representative on the board or in the Company Assembly is on her own. Secondly, the representation has to be demanded by the employees/ trade unions.14 Both the Basic Agreement and the Working Environment Act presume

12 Kings Bay refers to a mining accident at Svalbard in 1962. One of the questions that were raised was whether or not the accident could have been avoided if employee voice on H&S-issues had been louder, and this critique was supported by the conservative parties. Employee representation on the boards in state-owned companies introduced in 1965 was one of the results of the accident. Another was the fall of the Labour Party government. 13 But – it is important to add that previous studies (Hagen, 2010) show a great deal of “blending” between the two systems of participation; the trade union representatives ensure the board positions and use this role power in order to achieve influence as trade unionists (and vice versa). Our findings are however limited to the board position; the Company Assembly is not perceived as important enough to “spend” an important trade union rep (as the law prevents the same person serving both at the assembly and the board). The importance of the board triumphs the assembly. 14 Trade unions covering two-thirds of the employees may demand representation, and they may demand certain electoral procedures. This is the only legal role of the trade unions.

84 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

a managerial duty to inform and consult the representatives of the employees15. The BLER, in contrast, may be labelled “democracy by demand”. Another deviance from the negotiation system is the existence of a threshold of 200 employees (or 30 concerning board representation); there is no threshold for the right to information and consultation in the Basic Agreement. Thirdly, the opposite part in the collaboration system, the CEO, is the agent of the board; he is holding a subordinate position. The status of the representative(s) is also somewhat different; they have the same legal responsibility and rights as the directors elected by shareholders but are always in a minority position. When collaborating with management in line with the agreements, employee representatives are subject to the managerial prerogative. Finally, the question of whether employee representatives are accountable to employee interests or to company interests is still an unsolved issue. This was one of the conflict areas between the Labour Party and the LO in the 1960s, the party emphasized company interests while the LO emphasized the representative role – in line with the role of trade union representatives. Attending company interests also included a responsibility to ensure production growth and welfare and societal influence.

5.2.1 Company Assembly and Corporate Governance Searching for arguments supporting the assembly as part of the governance structure of the company has been an almost fruitless task (see Hagen, 2014a). Problems with including a new company body – with new tasks and responsibilities – were hardly touched in the debate. But in fact, the introduction of the assembly changes the governance structure. The “relevant” object of property in the context of the private company is the share. The right to speak and vote at the General Assembly and the right to receive dividends are the privileges given to shareholders. The Norwegian General Assembly has, in a comparative perspective, a high degree of authority and may decide on any matter, as long as the decision is not legally placed elsewhere (e.g. with the board or with the CEO). A very important, if not the most important (Hagen, 2014b), task of the General Assembly is to elect the board – which is how shareholders exercise their ownership. The board is responsible and must ensure that the company is run in accordance with the best interests of the company. The Company Assembly arrangement limits the privileges of the shareholders in both areas: some of the privileges of the General

15 In case of no trade union being present at the company, the employer has to inform and consult representatives of the employees (Chapter 8 in the Working Environment Act), these representatives must be found. And likewise – the employer need to make sure that both H&S Officers and members if the Working Environment Committee are elected.



”The Democratic Company Assembly” in 2014 

 85

Assembly are transferred to the Company Assembly as board members are elected by the Company Assembly – that is, indirectly by an assembly where shareholders are entitled to elect only two-thirds of the participants. The second arrangement – the right to elect employee representatives to the board – prevents the shareholders from composing the board of their own choice. In smaller companies (30-199 employees) this has to be demanded by the employees/trade unions, but in companies obliged to set up a company assembly, board representation is also mandatory. Nevertheless, in both company bodies the majority of participants are elected by the shareholders. Whether or not the employee representatives are able to exercise any influence is beyond the scope of this chapter (see Hagen, 2010); the important issue here is simply that the arrangement does limit the shareholders’ possibility to have complete control over the company bodies.

5.3 ”The Democratic Company Assembly”16 in 2014 No precise records of the actual extension of Company Assemblies and BLER exist. This is partly due to the fact that representation is not automatic, and there is no governmental body to ensure that representation is established. The trade unions give priority to the Basic Agreement and the social dialogue between the managerial prerogative and organized labour, and it could very well be argued that the rise in Corporate Governance and “shareholder value” has not been met with a corresponding interest in the possible regulation of shareholders at the company level (Hagen, 2011). Several historians (see Bergh, 1983, 2009) have argued there was little or no urge at the shop floor for the privileges of representation. However, these explanations only help us understand why the arrangements have fallen into the shadow of the collective agreements; it does not shed light on why the two different arrangements based on company law met with such an unequal fate.

5.3.1 Coverage 1: Company Assemblies Company Assemblies or their members are not, unlike the board members and CEOs, registered in the official Norwegian business register (“Brønnøysundsregistret”). However, existing information on all companies with more than 30 employees and all groups registered in Norway indicates that there are 991 companies and 892 groups with more than 200 employees in Norway. Data have been collected by telephone interviews with CEO and board chairs covering 230 companies and

16 The expression was used in the Labour Party program in 1969. www.nsd.uib.no/polsys/data/filer/ parti/10049.rtf).

86 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

117 groups (on the sample17 see Hagen, 2014, forthcoming). Preliminary results confirmed nine assemblies in the first group (2,5 per cent) and eight (7 per cent) in the latter. An extensive Internet search was also conducted18 with limited success. Even if the research may be subject to methodological criticism, it is highly certain that no assemblies have been left out.19 Institutionalized practices have three indicators: “they are widely followed, without debate and exhibit performance” (Tolbert & Zucker, 1983, p. 25). Dissolving the assembly is a common practice; it is done without debate. As a former employee representative in one of the largest Norwegian companies told us: When the company was set up [by a fission process], the management wanted to be very international. Industrial democracy arrangements were perceived as “too Norwegian”.

When changes in organizational design and company form lead to the establishment of the board as a company body, as for example in the large hospital reform in Norway in 2001, seats are always reserved for the employees, but no Company Assembly is set up. Recent cases of fission liquidating the assembly are the establishment of Yara, formerly a part of Norsk Hydro, or the splitting of Borregaard from Orkla.

5.3.2 Coverage 2: Board members Employees are entitled to demand representation both at the company and group levels. Table 1 presents the percentage of companies (including parents20) and company groups where the employees are represented on the board. The different categories refer to the different thresholds in the Companies Acts.

17 In short – companies with more than 200 employees and WITH board representation were selected. Among the group companies (parent companies) we chose only companies with three or more board members in order to avoid holding companies. 18 Norway has, as have most countries, established a Corporate Governance code. The governing principle of these codes is “comply or explain”; for example, the companies have to inform the (possible) shareholders and the public on their governance structure. Most large companies have thus included “Corporate Governance” or “Investor Relations” in their website, which have made company information more accessible. 19 However, we should add that the list consists of several of the most important companies in Norway (e.g. Statoil, Telenor and Norsk Hydro). And very interesting in its own right, the list of members of the assemblies is a regular ‘Who is Who’ of Norwegian business life. 20 Parent companies are also companies in their own right and thus subject to the same regulation. The BLER in a parent company could either represent the employees in the company only or all employees in the group.



”The Democratic Company Assembly” in 2014 

 87

Table 5.1: Percentage of companies with employee representatives at the board Number of employees

Companies (incl. parent)

Parent only

30-49 employees (one rep by demand) 50-199 employees (one-third by demand) 200+ employees (one-third mandatory) Total N

12 31 57 26 6,306

24 45 73 45 2,583

Note: Only groups with at least three board members and one employee in the parent company are included in order to avoid “empty companies”.

At present, only 26 per cent of company boards and 45 per cent of parent boards have employee representatives. Size is crucial (see Hagen, 2011 for a full analysis on variables); as shown, among the largest companies the employees are represented in almost half of the companies and 59 per cent of the groups. More than 40 years after the reform was passed a substantial proportion of companies and groups have no employee reps, even though representation is mandatory. An important question is whether these figures make it possible to label the arrangement a success compared to the Company Assembly. Before conclusions are drawn some important factors should be added. First, there may still be some errors in the register, we do know that some companies only register board members and fail to indicate whether they are elected by the shareholders or by the employees.21 Secondly, examining the structure of groups is very complicated; there are a lot of rubber stamp boards and “empty” companies, both at subsidiary level and at the top of the groups. Thirdly, local trade unions take into account the benefits of demanding representation; if the board is considered less important as a part of the decision- making process in the company, they refrain from making the demand. Sometimes the board in (one of) the subsidiaries is the most important arena, sometimes the parent board. Thus, an employee in a group might be “covered” by the agreement even if no representatives are recorded in the company where she has her legal contract. An important conclusion is that ownership matters, but basically as a factor (un)empowering the board and not due to shareholder resistance to representation (see Hagen, 2011 on the “non-use” of the arrangement). All in all, if the criterion is that “employees are present on the decision-making board” we expect

21 Prior to the gender reform (40 per cent of each gender in public limited companies [ASA]) no distinction was made between the two groups. The reform implied that quotas are relevant in both groups, and the need for registration emerged. But, the vast number of companies is limited companies (AS), and it seems likely that some continued the old way. And secondly – it seems less likely that shareholder-elected members (the “normal”) are marked “elected by the employees” than the opposite.

88 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

the figures to rise. Moreover, even if most companies in Norway are small, the large ones employ the majority of the work force and the larger the company, the more likely it is that representatives are found. 5.3.3 Contested Arrangements? 22 Coverage is one way of measuring success; another is whether the intentions of the reform have visible results today. The level of employee influence and co-determination, as well as the possible contribution to productivity and welfare, is well established in Norway (NOU 2010:1, Chapter 9). The literature on the effect of the institutional arrangement is huge. Concerning the BLER in particular, earlier results confirm that little or none of the old employer side resistance is left (Gulbrandsen et al., 2002). Likewise, among trade unions the old ideological scepticism towards the arrangement is hardly found, even if collective agreements are still ranked higher among the possible participation tools (Hagen, 2011). But, as shown in the most recent green paper on working life participation (NOU 2010:1, Chapter 10), there is sturdy resistance among employers to any extension of the arrangement; a suggestion from the labour side in the committee to lower the threshold from 30 to 10 employees was point-blank rejected. Turning to the Company Assembly, the dramatic decrease is confirmed: If the arrangement [the Company Assembly] is maintained, it needs things to do. In the present situation it makes no sense at all. You would have to transfer some of the tasks of the board. I wouldn’t necessarily recommend that, but if you do have a Company Assembly there has to be some meaning in it. (Shareholder rep/CEO)

The Company Assembly is rejected both from a corporate governance perspective and as an instrument in the hands of the employees. For them the assembly is basically an asset to use as part of a trade-off: From an employee perspective – everybody agrees on the fact the board is the key instrument. Employee interests are best served from a board position. It is better to expand the representation [cf. the legal requirement of an agreement]; any influence on production processes and organizational design comes from participation on the board. (Shareholder rep/CEO)

22 This section is based on an ongoing project at Fafo (see Hagen 2014a for the first part of this project – the Company Assembly seen from a Corporate Governance perspective). In total 25 people were interviewed. All have/or recently had positions as board chairs/members, CEOs and chairs/members of a company assembly. All came from well-known large companies. Three were experienced company lawyers. We have also interviewed representatives of the social partners at the central level.



Several Attempts – One Success? 

 89

The Company Assembly is out of date and has no function as an operative body. It has however one important function – the employees may, as a quid pro quo, expand their representation on the board. (Employee rep)

Several arguments are used to explain the lack of relevance of the assembly: timing, accountability and transparency and confidentiality. The proceedings at the assembly are not confidential; this seriously limits the possibilities to discuss strategy prior to decisions. The Company Assembly does not comply with modern norms of Corporate Governance (see Hagen, 2014b). Increased shareholder leeway due to both liberalization and “shareholder value” has made the board more important as a company body but has left little or no room for this Norwegian intermediary.

5.4 Several Attempts – One Success? While the assembly is considered an irrelevant, old-fashioned nuisance, the board is becoming more and more important (cf. the emphasis on the board in the different Corporate Governance codes). One question is whether the Company Assembly failed to connect – failed to fit in with the rest of the arrangements which constitute the institutions of collaboration? Another question is whether the arrangement failed to establish itself as an institution. Thelen (2009, p. 491) points out that building institutions is often a matter of political compromises; “institutions and rules are often ambiguous from the beginning, almost by design”. In addition to the double aim of democracy and productivity (Bergh, 1994), in our context there is another important duplicity: The reforms were aimed at both individual rights for workers to influence their own life and working condition and simultaneously to modernize earlier notions of a socialist society by increased influence from society on company decisions. The Aspengren Committee in 1962 stated that We need to attack the fundamental pattern; property rights of capital are the sole foundation for ownership of company surplus. There is partly a need for a new organizational form – both in order to democratize the working life and to increase productivity. The Company Assembly should be made up of representatives of the employees, the capital owners and stakeholders/ bodies with a direct interest in the well-being of the company. Local authorities belong to the third group. (after Jacobsen 1977, p. 14, my translation)

The Committee points out that an arrangement with a majority of representatives elected by employees and local authorities either would presuppose changes in the Constitution or alternatively a large expropriation programme. It is interesting to note how Aspengren comments on this issue: “Constitutional change demands a larger majority in the Parliament than the labour movement ever has been able to mobilize” (Jacobsen, 1977, p. 15, my translation). An expropriation programme is dismissed because it would destroy the state economically. Aspengren also places his worries at the local level, on possible consequences in the companies. Management obstruction

90 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

might occur and thus threaten productivity and profitability of the company, “which are also realties to consider” (ibid.). These arguments might be read as a combination of a wishful revolutionary dream, well-known pragmatism from trade unions and familiar concerns about productivity. But they may also be read as a sign of a still unresolved question of regulation of property rights in the trade union movement. The last confrontation over property rights took place in 1980. A committee appointed by the LO and the Labour Party (“Skytøen-utvalget”) argued both to strengthen the rights of employees and to strengthen the role of public authorities in the business sector. They proposed to expand the proportion of employee elected directors from one-third to 50/50. Moreover, they recommended to grant local governments the right to elect observers on the board (the right to speak and make suggestions, but no voting privileges), in line with a proposal from another LO committee (“Laake-utvalget”) concerning the Company Assembly in 1977. A short but loud public debate followed. Any expansion of representative rights was bluntly rejected by the employer side. Parity – even with a casting vote for the chair – was perceived as a violation of private property rights. Legal arguments were actively used with reference to Article 105 in the Constitution. Bergh (1983) argues that the proposal was interpreted by the employers as a part of a nationalization strategy, and this might explain their harsh resistance. The suggested inclusion of “public observers” is perceived as a sign of a still unresolved relationship between the role of society and the role of the employees in developing and expanding a democratic working life (ibid.). Curbing property rights by regulating the power balance at the company level seems to be a very ambitious idea. The ideas of the Skytøen Committee failed, and as history has shown, neither has the assembly brought any societal influence on company decisions today. Two other attempts may also be mentioned.

5.4.1 1920 and 1945 The first attempt to include workers in the decision-making of the company was made in a law passed in 1920, allowing workers to demand the establishment of a “Worker Committee” (“arbeiderutvalg”) in companies with more than 50 employees. The committees could request information and consultations; moreover, it was a managerial duty to include the committee in decision-making on certain issues. The arrangement was not very successful; very few committees were actually established. Jakhelln23 discusses two different explanations. One is the hardship of unemployment and deflation in the 1920s and 30s. More important in our context is that the work of

23 http://snl.no/bedriftsdemokrati. Read Oct. 10, 2014.



Several Attempts – One Success? 

 91

the trade union representatives made the committees redundant. Even so, the law was not annulled until 1963. The second attempt was the Production Councils in 1945 (see Falkum, this volume). The councils were conceived as part of a corporative structure made up on the national (“samordningsråd”), sectorial (“bransjerådene”) and company (“produksjonsutvalgene”) levels. The national council – where capital, labour and consumer interests were represented – was meant to be an advisory council for the government on economic policy. Due to conflicting views among the participants the committee never lived up to its planned importance, and the council was dismantled in 1954. Likewise, the sectorial councils did not fulfil the intentions of the social partners to agree on sectorial development and restructuring in different industries (Falkum, 2008). My concern is however with the third part: the local Production Councils. Both employers and employees were to be represented, and the aim was to set production targets and discuss restructuring and production growth in the individual companies. The councils were included in the Basic Agreement, and Falkum (2008) draws the line from the 1920 legislation – cooperation and co-determination should also influence the distribution of surplus. The Secretary General of the Labour Party and an important force behind the corporative programme, Haakon Lie, wrote in 1946 that The councils may truly contribute to the overall goal of the labour movement; a society with increased welfare for all, which provides individuals the opportunity to prosper in accordance with their own abilities, and provides the workers of the hand and of the mind the right to participate and make decisions in economic life as well. (after Falkum, 2008, p. 251, my translation)

Lie, as Falkum (ibid.) points out, looked upon the councils partly as a path to economic democracy and partly as an economic tool for economic growth. Approximately 1000 councils were established – a significant, but relatively low number. Several explanations have been mentioned (Bergh 1987); the content of the consultations did not meet the requirements, and the employer side was often sceptical, but scepticism was also considerable among the trade unions. Falkum (2008) draws the same conclusion: Faith in the councils was insufficient, in line with the two other building blocks of the corporative system of 1945. The four examples mentioned – the Company Assembly, the Skytøen Committee, the 1920 committees and the Production Councils – have one important feature in common: they all are perceived as tools for economic democracy at the societal level and increased social influence on company matters. By changing the power relations at the local level – inside the company – property rights would become less important. However, the attempt to “socialize” from below failed.

92 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

5.5 Institutional Change and Consistence 5.5.1 Effects of Legal Changes The legal changes in 1972 were introduced to increase employee influence on company matters. By restricting shareholder power at the local level, society would take one step further towards a democratic society in the economic area. Thus, our first question concerns whether the new arrangements represented a significant institutional development in that direction. A simple “no” seems appropriate; the division between economic and industrial democracy – between regulation at the national level and the power relations at the local level – is still present. The social partners are still the major players, even if ownership and shareholding is more important in contemporary Norway than in the 1960s. Both the Company Assembly and BLER were “alien” to the logic of the collaboration model. The next question is why only one of them was able to institutionalize in its own right. There is one obvious difference between employee representatives at the board and the Company Assembly: It is hard to picture limited companies without a board, but it is possible to conceive a company structure without the assembly. Employee representatives on the board take part in an established body, whereas employee representation in the assembly is the sole reason for the assembly’s existence. Thus, despite their relatively marginal position, the BLER are part of a well-known governance structure based on the notion of shareholder democracy and property rights. The minority position of the employees makes it possible for shareholders to carry on as usual. However, BLER is an institutionalized practice, and the elected employees play an important role on the board (Hagen, 2010). The need to combine democratic rights and company productivity is inherent in all the different elements in the model – for example, in the Working Environment Act as well as in the Basic Agreement. The fundamental conflict has nevertheless not disappeared. Owners want surplus and dividends; workers want reasonable working conditions, just pay and a say in important decisions. As long as the system is able to fulfil these expectations we can expect the system to prevail. Different parts, as in our case the representation rights prescribed by the Companies Act, must attend to both interests; or more crudely, actors with different interests must be able to answer “What’s in it for me?” If non-shareholder interests are the sole reason for establishing the Company Assembly, we might expect justification on behalf of the assembly to exceed the need for the board representatives. It should either make a difference to the employees or their constituency or contribute to increased productivity and company surplus. None of these concerns are mentioned in the above interviews. Respondents did not set the main focus on representation but on the arrangement as such; the Company Assembly does not make any important decisions and is a “rubber stamp” arrangement. In line



Institutional Change and Consistence  

 93

with the argument cited above, the assembly does not fit into the institutionalized company structure. Being a part of this structure, employee representatives on the board should also make contributions. From a shareholder perspective they contribute to productivity and growth. The interviews confirm that enthusiasm for democracy is hard to find among CEOs and members elected by shareholders; the core of the matter is shareholder interests. The board agenda became more important, and not less, as sometimes predicted, after the employees entered the boardroom. The findings support our hypothesis: Unless conflicting interests are able to benefit from the arrangement, the mismatch with the overall framework will eventually lead to changes; hence, the Company Assembly could be expected to disappear completely.

5.5.2 Layering Layering and drift are two different forms of institutional change. When examining the fate of the changes in company law, layering seems like the most relevant way of looking at the reform. If important decisions are made, the employees need to be informed and consulted; this is the logic of the Basic Agreements.24 Involvement on the board – where important decisions are made – needed to be included in the system. Because important decisions are not made in the Company Assembly, this element failed. The arrangements might have been intended to start a new way of regulating Norwegian capitalism, but as such no institutional changes has taken place. The employee reps at the boards are representatives of the employees, not of society, neither at the local or national levels. Disagreement inside the labour movement and political circumstances can explain why the last element ended up in a somewhat different shape than the rest of the pieces – as mentioned: with a legal basis and a right of the employees, not the trade unions. And as such, potential for drift was and is present: One of the counterarguments was that the BLER might reduce the position of the trade union reps in the company and thus weaken the core of the institution. We have no indication of this in our material, but if the legal possibility of including foreign workers in the BLER arrangement at the group level becomes more common (see Hagen & Mulder, 2014), this might weaken the Norwegian trade union reps and generate changes in the social dialogue at the company level. The distinction between strong and weak veto possibilities (Mahoney & Thelen, 2010, p. 19) is well suited for analyzing the importance of BLER. The ongoing debate over employee versus company (read shareholder) interests has yet to establish a

24 Cf. “Before adopting any decision on matters that concern the employees’ jobs and working conditions, this shall be discussed with the shop stewards” (Basic Agreement LO-NHO 2010-2013,§9-6).

94 

 Participation and Co-Determination: Why Some Arrangements Fail and Others Prevail

transformation from employee representatives to company stewards; the strength of the trade unions and the institutionalized norms of employee participation has “vetoed” the drift of the arrangement. We have, however, met employee reps in nonorganized companies committed to shareholder value only (see Falkum et al., 2009). This means that collective agreement is not only important for the establishment of BLER arrangements but also for the way they are institutionalized in the individual company (see Hagen, 2014a for the extended argument). The heated debate prior to the reform in 1972 has no corresponding sequel, even if several occasions have been inviting, particularly the large reform of the Companies Act that took place in 1996, after years of preparation. A more recent example is the already mentioned green paper committee on participation from 2010. There, too, the assembly was taken for granted and was not touched upon; even though all relevant positions were represented in the committee, the core of the matter was not processed (NOU 2010:1, Chapter 10). Even as a sleeping arrangement the Company Assembly has two ways of affecting the system. As mentioned, it is used in a trade-off for increased representation on the board. Secondly, the existence of the legal requirements and the perception of history is important. All employee rights, however unused, represent small or large victories in the struggle of the trade unions; the rights became institutionalized. This is well illustrated by one of our respondents, a very experienced company lawyer, who in his reflections on the future of the failed arrangement stated the obvious in an institutionalized Norwegian setting: A debate on the [future of] the Company Assembly might open up “the whole debate” [on participation arrangements] between the LO and NHO. The NHO opposes the arrangement, but they are probably very cautious towards the LO. The LO has the strong hand. The NHO has to give, and they probably won’t.

In other words, institutional mismatches may continue – at least as sleeping provisions – when powerful actors have vested interests.



References  

 95

References Bergh, T. (1983). Medbestemmelse eller opposisjon? In T. Bergh (Ed), Deltakerdemokratiet (pp. 84-125). Oslo: Universitetsforlaget. Bergh, T. (1987). Storhetstid (1945-65). Arbeiderbevegelsens historie i Norge. Oslo: Tiden. Bergh, T. (1994). Medbestemmelse som etterkrigshistorisk forskningsfelt. Bergen: LOS. Bergh, T. (2009). Kollektiv fornuft. LOs historie, bind 3. Oslo: Pax. Engelstad, F., Svalund, J., Hagen, I. M., & Storvik, A. E. (2003). Makt og demokrati i arbeidslivet. Oslo: Gyldendal Akademisk. Falkum, E. (2008). Makt og opposisjon i norsk arbeidsliv. Oslo: Fafo. Falkum, E., Hagen, I. M., & Trygstad, S. (2009). Bedriftsdemokratiets tilstand. Oslo: Fafo. Grønlie, T. (1977). Norsk industripolitikk 1945-65. In T. Bergh (Ed.), Vekst og velstand (pp. 99-166). Oslo: Universitetsforlaget. Gulbrandsen, T., F. Engelstad, T. B. Klausen, H. Skjeie, M. Teigen, & Ø. Osterud. (2002). Norske makteliter. Oslo: Gyldendal Akademisk. Hagen, I. M. (2010). Det mektige mindretallet. Ansatterepresentasjon i styret mellom Corporate Governance og Industrial Relations. Oslo: Fafo-rapport. Hagen, I. M. (2011). Medarbejdervalgte medlemmer av bestyrelsen – hvorfor benytter ikke alltid de ansatte deres rettigheter? Tidsskrift for Arbejdsliv, 1/2011, 47-63. Hagen, I. M. (2014). Board-level employee representatives in Norway, Sweden and Denmark: Differently powerless or equally important? In M. Hauptmeier & M. Vidal (Eds.), Comparative political economy of work. Critical perspectives on work and employment (pp. 139-161). Palgrave Macmillan. Hagen, I. M. (2014a). Bedriftsforsamlingen – pleie eller plikt? Oslo: Fafo. Hagen, I. M. (2014b). Company assembly – the forgotten or neglected part of industrial democracy? Paper to the NWLC 2014 in Gothenburg. Hagen, I. M. (2014 – forthcoming). Bedriftsforsamlingen i et bedriftsdemokratisk perspektiv. Oslo: Fafo. Hagen, I. M., & Mulder, B. J. (2014). Transnational employee representation on company board: The Scandinavian model. I Edoardo Ales & Iacopo Senatori (Eds.), Transnational dimension of labour relations. A new order in the making? (pp. 135-170). Torino: G. Gippichelli Editore. Heiret, J. (2003). Et nasjonalt system i en internasjonal verden. In J. Heiret, O. Korsnes, K. Venneslan & Ø. Bjørnson, Arbeidsliv, historie, samfunn. Norske arbeidslivsrelasjoner i historisk, sosiologisk og arbeidsrettslig perspektiv (pp. 177-246). Bergen: Fagbokforlaget. Jacobsen, A. (1977). Bedriftsdemokrati : forslag og tiltak: Tor Aspengren om ideer og erfaringer. Oslo: Tiden. Lindkjølen, K. M. S. (2009). Bedriftsforsamlingens beslutningskompetanse i aksjeselskaper og allmennaksjeselskaper : En rettsdogmatisk fremstilling med en arbeidsrettslig vinkel. Master Thesis, University of Oslo. Mahoney, J., & Thelen, K. (eds.) (2010). Explaning institutional change. Cambrigde: Cambridge University Press. NOU 2010:1 Medvirkning og medbestemmelse i arbeidslivet. Scott, W. R. (2008). Institutions and organizations. Ideas and interests. Los Angeles: SAGE. Sejersted, F. (1984). Opposisjon og posisjon: Høyres historie. Oslo: Cappelen. Sejersted, F. (2001). Struktur og legitimitet. Søkelys på arbeidsmarkedet, 7/2001. Sejersted, F. (2011). The age of social democracy. Princeton: Princeton University Press. Selznick, P. (1957). Leadership in administration. Berkeley: University of California Press. Thelen, K. (2009). Institutional Change in Advanced Political Economies. British Journal of Industrial Relations, 47(3), 471-498. London: Blackwell Publishing. Tolbert, P. S., & Zucker, L. G. (1983). Institutional Sources of Change in the Formal Structure of Organizations: The Diffusion of Civil Service Reform, 1880-1935. Administrative Science Quarterly 28 (pp 22-39).

Mari Teigen

6 The Making of Gender Quotas for Corporate Boards in Norway The gender quota reform for corporate boards, first adopted in Norway in 2003 and fully implemented from 2008, has had great repercussions. A wave of diffusion of corporate board quota legislation has swept across Europe, and some other parts of the world (Fagan et al., 2012; Teigen, 2012a and b; Armstrong & Walby, 2012; Terjesen & Lorenz, 2014).1 In Norway, as an effect of the reform, the presence of women in corporate boards has increased dramatically over the last 10 years – and gender balance is evolving in other countries, especially in those where gender balance policies have been introduced for corporate boards (EWL report, 2012). This chapter departs from the ongoing European processes of gender quotas for corporate boards being in the making, takes a step back and analyzes the starting point: the making of gender quotas for corporate boards in Norway. This reform constitutes an innovative, however at first controversial, legal regulation. No other country had prior to Norway even considered to regulate the gender composition of corporate boards. To some extent gender quotas for corporate boards are at odds with the Norwegian industrial relation system, with its relatively clear boundaries for state interventions in the governing of private capital. Thus on the one hand, the introduction of gender quotas for corporate boards breaks with the established institutionalization of relations between employers, employees and the state. On the other hand, gender quotas for corporate boards constitute an enlargement of the relatively strong gender equality institution in Norway. At heart of this chapter is the question of why the gender equality institution in this case trumped the established boundaries for state intervention in the governing of private capital. The purpose is to describe and explore the interplay of contextual factors and processes that in sum pushed towards the making of the gender quotas for corporate boards. The chapter addresses national preconditions and processes and addresses questions about how this reform fits with the Norwegian gender equality institution: what factors facilitated the policy process and what was the role of political agency? Theoretically the chapter lends itself to insights from historical institutionalism (Thelen, 1999, 2014), including perspectives of path dependency (Pierson, 2004) and gendered institutional analysis/feminist institutionalism (Waylen, 2014; Krizsnan et al., 2012; Mc Bride & Mazur, 2010). In the next section I present some theoretical perspectives mainly belonging to the historical institutional theory tradition. Subsequently I provide an account

1 Spain (2007), Iceland (2009), France, Belgium, the Netherlands, Italy (2011) and Germany (2014) have adopted gender quotas for corporate boards. © 2015 Mari Teigen This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Theoretical Perspectives 

 97

of the Norwegian legislation for gender quotas on corporate boards and show the development of gender balance on corporate boards over the last decade. The major part of the chapter presents an analysis of the main factors and processes in the making of gender quotas for corporate boards in Norway. In the final part I discuss how these different factors and processes interplayed in relation to three central topics in historical institutionalism: temporality, path dependency and policy agency. I conclude by reflecting upon the development of gender quotas for corporate boards in light of political quotas and thus close with a larger discussion about the demarcation of the political and the economic sphere.

6.1 Theoretical Perspectives Historical institutionalism offers approaches to investigate how and why institutional change occurs (Thelen, 2003; Mahoney & Thelen, 2010). Feminist institutionalism is a variant of historical institutionalism particularly concerned with gender outcome and institutional continuity and change in transitions to state feminism and institutionalizations of gender equality policy (Waylen, 2014). Discursive institutionalism emphasizes the role of ideas and discourse in politics (Schmidt, 2008). Historical institutionalism helps explain patterns of events by taking processes in time – history – seriously. The historical institutional perspective directs attention to the relationship between sequences, trajectories and contextual factors of importance to explain social, political and economic change and stability (Mahoney & Thelen, 2010). Stability and perpetuation are the basic traits of institutions and a main contribution of institutional theory concerns its exploration of mechanisms of stability. The fact that institutions are not always stable has been explained as the result of exogenous shocks and critical junctures. A limitation of this perspective is however that change becomes too closely tied to abrupt transitions. In contrast, Mahoney and Thelen (2010, pp. 5-7) call for a general model of institutional change that takes into account both exogenous and endogenous sources of change. They draw a framework for institutional change consisting of the political and institutional context as drivers of processes in interaction with change agents. The policy context, and in particular the notion of path dependency, is central for understanding the emergence and particular formulation of policy reforms. The notion of path dependence is often used to describe why a certain development is chosen over another. The emergence of a particular institutional arrangement becomes decisive for later reforms. Path dependence has, however, been defined by different degrees of precision – in a broad or a narrow sense, in a loose or a rigid way. Paul Pierson comments that as a consequence of the lack of rigidity we may end up with a concept lacking a clear meaning (2004, p. 10). In a socio-historical version path-dependent processes emerge and are embedded in concrete temporal processes interconnected

98 

 The Making of Gender Quotas for Corporate Boards in Norway

with traditions, norms and expectations which drive processes in certain directions, while alternative routes would demand more thorough justifications and persuasion. Still, the main contribution of the notion of path dependence is the insistence on the significance of the order of events – on sequences and the temporal dimension of social processes (Pierson, 2000, 2004). Calling attention to the temporal context is important for understanding policy processes, as well as the role of policy agency. In this chapter the notion of path dependency is used to underscore that institutional and policy legacies are important to determine both change and continuity. The main point is to argue that there exists a policy legacy in Norway for solving “problems” of male dominance in decision-making by applying gender quota arrangements. The processes that led to the making of gender quotas for corporate boards in Norway are particularly interesting in regards to the way they bridge debates about gender balance in political and economic decision making. The institution studied in this chapter is the Norwegian gender the Norwegian gender equality policy institution, in particular, the regulation of gender balance in corporate boards. Simultaneously, this is to some extent also a study of the corporate board institution, where new regulations of gender representation have been included in the more general regulation of the corporate board institution.

6.2 A Note on the Data The analysis presented is based on all relevant documents from the political process, as well as other types of information, such as from the media debate. The most important documents are: 1) the consultation proposal on the revision of the Gender Equality Act, from the Ministry of Children and Family Affairs 19992, 2) the white paper from the Ministry of Children and Family Affairs: Proposition on reforms to the Gender Equality Act (2000-2001)3 3) the consultation proposal on gender representation in public limited companies, state limited companies and state businesses, etc. and the

2 Consultation proposal from the Ministry of Children and Family Affairs 1999. (Høring: Forslag til endringer i likestillingsloven, Barne og familiedepartementet, 1999). 3 http://www.regjeringen.no/nb/dep/bld/dok/regpubl/otprp/20002001/otprp-nr-77-2000-2001-. html?id=123306, (08.05.2013). Proposition to parliament 77 (2000-2001), Ministry of Children and Family Affairs: Ot.prp. 77 (2000-2001), Om lov om endringer i likestillingsloven mv., Barne- og familiedepartementet.



National Preconditions and Processes 

 99

proposal to change the Companies Act and some other acts4, 4) White paper from the Ministry of Children and Family Affairs: Proposition on reforms to company legislation on gender representation in company boards.5 In-depth analyses of the first stages of the political process are provided in Teigen (2002), Evenrud (2010), Engelstad (2011) and Sørensen (2011). Cvijanovic (2009) provides an analysis of the media debate on the issue of gender quotas for corporate boards. Central aspects of deregulation policies were described in the government commission report (NOU 2000:19) on the privatization of public enterprises.6

6.3 National Preconditions and Processes The Norwegian gender quota legislation for corporate boards applies to a wide range of companies: the boards of public limited companies (PLC), inter-municipal companies7 and state enterprises.8 Cooperative companies9 and municipal companies10 were included from 2008 and 2009. The numerous private limited companies, however, often small, family-owned businesses, were not subject to gender quota legislation. Nevertheless, the requirement of at least 40 per cent of each gender in the board rooms today regulates central parts of Norwegian business life. In this chapter the main focus is on public limited companies. The criteria for gender representation on the boards are set in the Norwegian Public Limited Liability Companies Act11 in its article 6-11a. Demand for representation

4 http://www.regjeringen.no/en/dokumentarkiv/Regjeringen-Stoltenberg-I/bfd/Horinger/2001/ Horing-kjonnsrepresentasjon-i-styrer.html?id=421560, (08.05.2013). Consultation proposal from the Ministry of Children and Family Affairs 2001. (Høring: Kjønnsrepresentasjon i styret i allmennaksjeselskaper, statsaksjeselskaper og statsforetak, m.v. – forslag til endringer i allmennaksjeloven og i enkelte andre lover, Barne- og familiedepartementet, 2001). 5 http://www.regjeringen.no/en/dep/bld/dok/regpubl/otprp/20022003/otprp-nr-97-2002-2003-. html?id=127203, (08.05.2013). Proposition to parliament 97 (2002-2003), Ministry of Children and Family Affairs: Ot.prp. 97 (2002-2003), Om lov om endringer i lov 13. juni 1997 nr. 44 om aksjeselskaper, lov 13. juni 1997 nr. 45 om allmennaksjeselskaper og i enkelte andre lover (likestilling i styrer i statsaksjeselskaper, statsforetak, allmennaksjeselskaper mv.), Barne- og familiedepartementet 6 http://www.regjeringen.no/en/dep/fad/documents/nouer/2000/nou-2000-19.html?id=117394, no translation to English. (08.05.2013) 7 http://www.lovdata.no/all/hl-19990129-006.html, article 10, no translation to English. (08.05.2013) 8 http://www.lovdata.no/all/nl-19910830-071.html, article 19, no translation to English (08.05.2013) 9 http://www.ub.uio.no/ujur/ulovdata/lov-20070629-081-eng.pdf, article 6-4a (08.05.2013) 10 http://www.lovdata.no/all/hl-19920925-107.html#80a, article 80a, no translation to English (08.05.2013) 11 The rules regarding representation of both sexes are to be applied separately to employee-elected and shareholder-elected representatives in order to ensure independent election processes.

100 

 The Making of Gender Quotas for Corporate Boards in Norway

of both genders on the board.12 Similar formulations of the legal demands apply for all types of companies object to the gender quota ruling. Figure 1 shows the increase in the presence of women on the boards of public limited companies. The legislation adopted in 2003 was formulated as a “threat»: if the companies did not voluntarily reach the gender demands by July 2005, the legislation would be effectuated. Although the representation of women increased between 2003 and 2005, the target set of at least 40 per cent women was far from reached by the public limited company boards. Thus, in late autumn 2005 the government decided to effectuate the legislation for newly established companies starting in 2006 and all PLCs starting in 2008.13 80

60

36

40

40

39

40

41

41

2009

2010

2011

2012

2013

25 18

20 6

9

12

0 2002

2004

2005

2006

2007

2008

Figure 6.1. Proportion of women on the boards of Public Limited Companies, Norway. 2002 – 2013. Source: Statistics Norway.

12 The demand for gender representation is formulated as follows: 1) Where there are two or three members of the board, both genders should be represented. 2) Where there are four or five members of the board, both genders should be represented by at least two members. 3) Where there are six to eight members of the board, both genders should be represented by at least three members. 4) Where there are nine or more members of the board, the membership should comprise at least 40% men and 40% women. 5) Rules 1 to 4 also apply for the election of deputy members. 13 A main reason for the effective implementation of gender quotas for corporate boards was probably the rather tough sanctions that were implemented for breaching the law. The Company Act applies identical sanctions for breaching all its rules, with forced dissolution as the final step. Consequently, a company that does not have a legal board, despite several warnings with the possibility of correcting the matter, will be subject to forced dissolution.



National Preconditions and Processes 

 101

6.3.1 The Norwegian Gender Equality Policy Institution The politics of state feminism (Hernes, 1987) in Norway can be characterized as consisting of three core elements. The first element is the Gender Equality Act, which was passed in 1979. The Norwegian Gender Equality Act combines protection against discrimination with active duties for public authorities and employers to promote equality. Furthermore, it has included provision for positive action from the beginning, which has paved the way for preferential treatment and gender quota arrangements. A second element is family- and welfare policies. The Norwegian welfare state adopted maternity leave schemes and sponsored childcare policies from early on, and these policies have been regarded as part and parcel of woman-friendly policies, characterizing Norwegian state feminism. The generous leave schemes and the extensive childcare provisions have facilitated the reconciliation between family and work by promoting women’s participation in the labour market, as well as promoting men’s participation in the family and care for children. Third and finally, gender quota policies were established in the 1970s in Norway to promote and regulate the gender balance of decision-making assemblies. In recent years gender quotas in politics have spread globally (Dahlerup & Freidenval, 2005; Krook, 2009; Franceschet & Piscopo, 2013).

6.3.2 Positive Action and Gender Quota Arrangements in Norway The existing positive action and quota arrangements can be divided into three main types: preferential treatment, promoting procedures and minimum representation rules. Preferential treatment is the most widely dispersed in recruitment and promotions in public administration (state and municipal sectors), in some private companies, and in connection with admission to gender-skewed types of education; applicants from the under-represented gender are given priority, when qualifications are equal or about equal. As a result of the careful formulation of these procedures, they have proven to have had only minor direct effects. Nevertheless, studies indicate that they to some extent positively affect organizations’ prioritizing and legitimizing of gender equality (Teigen, 2002). Promoting procedures constitute a slightly different kind of positive action procedure, which implies that candidates’ chances are improved by being moved upwards in ranking. The “additional point” system is the most commonly applied procedure to balance the gender composition of students within gender-skewed fields of learning. “Earmarking” is another promoting procedure, mainly applied at universities to increase the representation of women in academic positions. “Earmarking” of university positions for women, recruitment positions, as well as professor positions played an important role in the gender-equality strategy of Norwegian universities in the 1990s. This procedure was abolished after the procedure had been judged by the European Free Trade Association’s (EFTA) surveillance

Preferential treatment

Minimum representation

Minimum representation

Minimum representation

Minimum representation

Employment

Politics

Organizations

Public commissions

Corporate boards

Promoting procedures

Adopted in 2003 in company legislation (PLC), implemented for new companies in 2006, full implementation in 2008

1981, first regulation of gender composition in the Gender Equality Act, 1988, 40% of each gender, included in the Local Government act, 1992

The Norwegian Confederation of Trade Unions, 2005

Liberal Party 1974; Socialist Left Party 1975; Centre Party 1979; Labour Party 1983; Christian Democratic Party 1989

State sector, 1981; municipal sector, 1985

Minimum 40% of each gender should be represented in the boards of public limited companies and publicly owned enterprises

Minimum 40% of each gender should be represented in publicly appointed boards, councils and committees

Quotas (at least 40% of each gender) regulate composition of decision-making bodies, as far as it is possible

Quotas (at least 40% of each gender) regulate party election lists and appointments within party organizations in five of the seven major political parties

In recruitment and promotions applicants of the underrepresented gender are given preferential treatment when qualifications are equal or about equal.

Upper secondary education, 1982; Norwegian School of Candidates of the under-represented gender are given Economics and Business Administration, 1980 priority in cases of equal qualifications (same amount of school points). Norwegian University of Science and Technology, 1981 (additional points), 1997 (earmarking procedure) Additional school points granted to applicants of the under-represented gender; earmarking of school places for candidates of the under-represented gender, with strict restrictions according to qualifications.

Preferential treatment

Education

Procedures

Adopted

Institutional fields Types

Table 6.1: Types of positive action/quota procedures according to social fields

102   The Making of Gender Quotas for Corporate Boards in Norway



National Preconditions and Processes 

 103

authority (ESA), which concluded that this particular formulation of positive action would be in contravention of the principles of equality and proportionality.14 Minimum representation, or quotas in the restricted sense of the word, establishes requirements for the gender composition in terms of fixed distributions. In the Norwegian context, such arrangements are generally formulated as a demand for at least 40 per cent of each gender. Principles of minimum representation exist as voluntary agreements in five of the major Norwegian political parties, exceptions being the Conservative Party and the right-wing Progress Party. Voluntary quota arrangements are also widely dispersed among civil society organizations. Legislative quotas regulate the gender composition of publicly appointed boards, councils and committees, and were first introduced in 1981 in Section 21 in the Gender Equality Act. For more than 20 years this was the only legislative quota arrangement in Norway, until the revision of the Company Act in 2003. Gender quota arrangements have played a central role in the institutionalization of Norwegian gender equality policy (Teigen, 2011; Skjeie & Teigen, 2012). These schemes have for the most been rather powerful in achieving their aim, although the wider ripple effects have been modest (Skjeie & Teigen, 2012, p. 60).

6.3.3 Employee Representation on Corporate Boards The idea that gender balance on corporate boards could be regulated did not emerge in the public debate before the late 1990s. Until then it was taken for granted that it was primarily the responsibility of the owners to select the board members they found the best suited to protect the company’s interests. However, the owners’ autonomy to choose board members was already restricted. With the revision of the Companies Act in 1972, workers in companies with more than 50 employees were given the right to elect up to one-third of the board members; and in companies with more than 200 employees, the company was obligated to set up an enterprise assembly, where one third of the members were elected by employees and two thirds by shareholders (Hagen, Engelstad, this volume). Employee representation on company boards expresses an adherence to workplace democracy, as well as an extended view of board members mandate to be broader in terms of representing the interests of the owners’ shares in the company. The enlargement of industrial democracy was presented as a reason for introducing gender quotas for corporate boards in the consultation document in relation to the preparation of a revision of the Gender Equality Act.

14 A similar Swedish arrangement was abolished by the European Court of Justice in the Abrahamson Case (C407/98), as a contravention of the EU Equal Treatment Directive Article 141 (4).

104 

 The Making of Gender Quotas for Corporate Boards in Norway

In Norway the content of democracy is constantly evolving. How we conceptualize democracy is changing through practice, and demands to and implementation of democratic arrangements come for on more new areas. This was the reasoning behind the introduction of industrial democracy at that time. It could be relevant to draw a line between the development of industrial democracy and gender representation in company boards. Although there was disagreement about the introduction of industrial democracy, it is today granted and fully accepted that employees should be represented in company boards. (Ministry of Children and Family Affairs, Consultation document, 1999, p. 58, my translation). 15

In the public debate gender representation was by some fronted as an extension of workplace democracy. The parallel made between employee and gender representation emphasize that there already exist restrictions to the autonomy of the owners in relation to the composition of company boards, something that may have contributed to prepare the ground for gender representation on company boards.

6.3.4 Deregulation Meets the Institutionalization of Gender Equality Policies The first initiative to introduce gender quota regulation for corporate boards came in the government’s consultation with the social partners and civil society organizations as part of the preparation of a revision of the Gender Equality Act in 1999.16 As a part of the revision process the government proposed to expand the scope of article 21 in the Gender Equality Act to include all company boards. A context for the proposition to expand the scope of § 21 was the ongoing processes of deregulation. Heavy deregulation of publicly owned businesses was effectuated in Norway from the 1980 and onwards (Engelstad et  al., 2003; Stjernø, 2005)17. To some extent, an unintended effect of the deregulation processes was that the scope of the existing gender quota regulation, in article 21 in the Norwegian Gender Equality Act, was narrowed. Thus, deregulated, but still publicly owned companies were no longer subject to the regulation of gender balance in public boards and commissions. On this background the government argued in its’ consultation document of 1999 and in its proposition to parliament in 2002 that a gender balance regulation limited to state

15 Høring: Forslag til endringer i likestillingsloven (Barne- og familiedepartementet). (Consultation: Proposition for revising the Gender Equality Act (Ministry of Children and Family Affairs)) 16 Høring: Forslag til endringer i likestillingsloven (Barne- og familiedepartementet). (Consultation: Proposition for revising the Gender Equality Act (Ministry of Children and Family Affairs)) 17 According to Vogel (1996) considerable confusion exists over the term ‘deregulation’. Deregulation can be described as “the reduction and elimination of government regulations” and liberalization as “the introduction of more competition within a market” (Vogel 1996:3).



National Preconditions and Processes 

 105

companies would change the conditions for competition between private and public owned companies.18 The government wishes to grant all companies equal conditions to compete, irrespective of whether they are public or private companies. Rules of minimum representation of both genders in the boards of private enterprises will give private enterprises equal conditions for competition as for public enterprises. (Ministry of Children and Family Affairs, consultation document, 1999, p. 58, m y translation).19

Important parts of Norwegian gender equality policies have been oriented towards state control, rules and regulations, while policies of deregulation have been reasoned on the grounds of the necessity to reduce government regulation and power of business and industry (Derthick & Quirck, 1985). A main argument for deregulation is that state intervention hampers economic growth and that the role of the state needs to be down-played to the benefit of stringent market criteria to secure distribution and wellbeing (NOU 2000:19; Jenson, 2010; Simon-Kumar, 2011). In this light, deregulation and gender equality policies appear to be inherently contradictory. However, exactly because of this contradiction the possibility of an expansion instead of confinement of the scope of the gender quota legislation was made salient. In addition, the fact that the Norwegian state is a prominent owner and actor in economic life, and in particular among the listed companies registered on the Oslo Stock Exchange, probably has strengthened the legitimacy for the state to take action. More than 40 percent of the values on the Oslo Stock Exchange is publicly, mainly state, owned.20 Moreover, as the debate on gender quotas for corporate boards arose, there were simultaneously other movements in the direction of expansion of the impact field of gender equality policies in working life. Until the revision of the Gender Equality Act in 2002, the pro-activity clause of the Act had been restricted to the public sector; starting in 2002 the duty to promote gender equality includes all employers as well as the social partners (Gender Equality Act, article 1a). Hence, at the turn of the millennium, mandatory gender quotas on corporate boards and the expanded proactivity clause can be interpreted as an expression of movements in the border area between the state and the economic field.

18 Høring: Forslag til endringer i likestillingsloven (Barne- og familiedepartementet). (Consultation: Proposition for revising the Gender Equality Act (Ministry of Children and Family Affairs)) and Ot. Prp. 97 (2002-2003) (Proposition to parliament 97 (2002-2003)). http://www.regjeringen.no/nb/dep/bld/ dok/regpubl/otprp/20022003/otprp-nr-97-2002-2003-/6.html?id=127240 19 Høring: Forslag til endringer i likestillingsloven (Barne- og familiedepartementet). (Consultation: Proposition for revising the Gender Equality Act (Ministry of Children and Family Affairs)) 20 http://vpsinfo.manamind.com/sectorstats/index.do?l=no

106 

 The Making of Gender Quotas for Corporate Boards in Norway

6.3.5 Women in Management The “women in management” debate was high on the public agenda in the 1990s in Norway, as well as in most of the industrialized world. To some extent the debate hit Norway particularly hard by the way it interfered with the national self-image of being particularly successful in affairs of gender equality. In Norway, as in the other Scandinavian countries, a discrepancy exists between a relatively balanced representation of men and women in political decision-making on the one hand, and strong vertical gender segregation in the labour market, and especially in business (Teigen & Wängnerud, 2009; Niskanen, 2011). The small number of women in top positions in business has been perceived as paradoxical in light of the gender equality achievements within other society areas. Some studies indicate in fact that the representation of women in top management in business and industry is higher in many European countries and the U.S. than in the Scandinavian countries (Birkelund & Sandnes, 2003). At the same time, male dominance in top positions in business is a general trend characterized by surprising similarity across countries.21 Norwegian data show that the business elite express more negative attitudes towards gender equality policies in comparison with the other elite groups (Skjeie & Teigen, 2003). The poor representation of women in senior positions in economic life combined with a lack of will within business to adopt initiatives to promote gender equality probably formed an important backdrop for why improving gender balance in the board rooms became a pressing political issue. Consequently, the long established and generally accepted demand for self-regulation of private businesses was challenged by a growing claim for state regulation to promote gender balance.

6.3.6 The Public Debate on Gender Quotas for Corporate Boards The institutionalization of party quotas and legal quotas for public commissions were certainly important for why the introduction of gender quotas for corporate boards appeared applicable. Nonetheless, even though gender quota policies have been more widely used in Norway than most other countries, they provoke opposition. The debate prior to the adoption of gender quotas for corporate boards was heated and polarized. The public and political debates over the gender quota for corporate boards have been studied through analysis of the media debate (Cvijanovic, 2009) and through analysis of the policy documents prepared by the government and the responses of organized interested in the consultative processes (Teigen, 2002; Evenrud, 2010; Engelstad, 2011; Sørensen, 2011).

21 http://20-first.com/wp-content/uploads/20-first-2014-Global-Gender-Balance-Scorecard.pdf



National Preconditions and Processes 

 107

The participants in the public debate on gender quotas for corporate boards were primarily people in top positions. The main voices against quotas for corporate boards were expressed by corporate managers and owners and representatives of employers’ organizations, while the supporters were represented mostly by politicians and highranking civil servants, as well as actors connected to the gender equality machinery. The supporting politicians came from a broad spectrum of political parties, mainly the Norwegian Labour Party, the Christian Democratic Party and parts of the Conservative Party. Opponents among politicians mainly represented the Progress Party and the Conservative Party, but there were few participants from the opponents’ side representing the political parties. The women’s organizations did not take an active part either in the newspaper debate or in the consultation process. The central arguments in the public debate represented a wide range of opinions. Still they can be divided into three main types of argument, about justice, utility and democracy. The justice argument: Proponents of gender quotas for corporate boards argued that gender balance in economic decision-making positions concerns issues of justice, often without specifying when and how “justice” is relevant in this context. Given that justice is a matter of redistribution of resources, the claim is that positive action/ gender quota is a necessary tool to achieve gender equality. The strong presence of men in Norwegian corporate boards and in economic decision-making in general, is then posited as unacceptable and as a possible indication of unfair treatment of women. The main counter argument based on justice mainly argued that regulation of the gender composition would not be fair. Recruitment to corporate boards should not be based on the gender of candidates. The owners should have the right to select the candidates they find most suitable to sit on the board. Gender quota regulations were considered illegitimate unequal treatment and as discrimination against men. The utility argument: Arguments about utility and profitability were particularly central as arguments in favour of gender quotas for corporate boards. The human capital oriented argument claims that since the total talent potential of a population is distributed fairly evenly between men and women, male dominance in corporate boards indicates an under-utilization of women’s skills. The government explicitly argued that more women in company boards would be “good for companies”: More women in company boards could add new resources, competences and impulses to Norwegian trade and industry. The ministry considers it to be a competitive advantage to get women into the board rooms. To exclude half of the competent workforce from the board room, expresses a bad utilization of resources and can weaken the quality. Enterprises that do not recruit women will have problems acquiring the best resources. As the public sector now has achieved a proportion of women at 40 percent in commissions, boards, etc., the quality is raised. (Ministry of Children and Family Affairs, consultation document, 1999, p. 58). 22

22 Høring: Forslag til endringer i likestillingsloven (Barne- og familiedepartementet). (Consultation: Proposition for revising the Gender Equality Act (Ministry of Children and Family Affairs))

108 

 The Making of Gender Quotas for Corporate Boards in Norway

This argument is particularly interesting in the way it uses the rhetoric of business life and thus appears in a way more unassailable. Nonetheless, the main counter arguments based on utility claim that gender quotas on corporate boards would lead to less competent women replacing more competent men. This argument is grounded on the opinion that not enough women have the relevant experience, and that the recruitment of qualified women needs to start earlier and further down the organizational hierarchy to create a pool of wellqualified women. Moreover, a related argument against the gender quota reform has been that the gender quota legislation would lead foreign investors to be less inclined to invest in Norwegian companies, with the consequence that Norwegian companies would lose competitive advantage. The debate on the effects of gender quotas for business performance is ongoing. Some analysts find a negative effect (Ahern & Dittmar, 2012; Matsa & Miller, 2013), while others find zero effect or a moderate positive effect for the relatively poorly performing companies (Dale-Olsen et al. 2013) The democracy argument: Arguments concerned with democracy were also central for the governments’ justifications of the gender quota reform. The argument was that gender balanced participation in economic decision-making is crucial for the Norwegian democracy, and in particular the importance of equal rights to participation on the boards of the biggest and most influential companies, where the state is often a major owner. In addition it was argued that regulation of gender balance of corporate boards implied an enlargement of industrial democracy, as described earlier in this chapter. A main counter argument concerned with democracy holds that gender quota regulations were incompatible with the principles of shareholder democracy. The claim has been that gender quota regulations would hinder owners’ democratic rights to recruit the candidates they find the most suited, as well as interfere with the election process. The argument is that the owners invest and risk their own money and should therefore have the right to decide who shall represent them on the board. The debate in advance of the gender quotas for corporate boards’ reform was, as mentioned, heated. The debate did not muster active participation among diverse actors; it was rather a top-down process. Civil society actors, such as the women’s organizations, mainly did not participate in the debate. The position of the women’s organizations has been weak throughout the last couple of decades, and this was probably not viewed as an issue with mobilizing potential. Gender quotas for corporate boards were mostly viewed as an issue concerning the elite levels of society. The elitist aspects may be important for why the debate suddenly disappeared from the public agenda. As soon as the quota legislation was implemented and the statistics showed that the requirement of at least 40 per cent of each gender represented on the boards were met, the debate more or less disappeared from the public eye.



National Preconditions and Processes 

 109

6.3.7 The Policy Agency Context The policy process lasted about a decade from when the first government motion was sent to the consultative bodies in 1999 until the full implementation of mandatory gender quotas in 2008. The first initiative to introduce a gender quota regulation for corporate boards23 came in the consultation audit regarding a major revision of the Gender Equality Act in 1999 by the minority Conservative-Centre government coalition (Bondevik I).24 The minister in charge was Valgerd Svarstad Haugland in the Ministry of Children and Family Affairs, from the Christian Democratic Party. The gender quota motion was withdrawn however, due to necessary legal clarification. A new motion was sent upon consultation by the Labour Party government (Stoltenberg I) in 2001.25 Some important changes can be witnessed in this proposition in comparison to the white paper from 1999. First the gender quota regulation was to be regulated through the Companies Act, and not as in the previous round through the Gender Equality Act. Second, the companies subject to the legislation were the public limited companies in addition to the public companies (state and inter-municipal companies), while the 1999 proposition included all companies irrespective of type of registration. Third, the gender target was changed from at least 25 per cent of each of the genders to 40 per cent. The final adoption of a quota ruling in the Companies Act was based on the motion from 2001 and presented to parliament by a new Conservative-Centre government coalition (Bondevik II)26 in 2003 (Proposition to parliament, Nr. 97 [2002-2003]).27 When the new gender quota law was passed in Parliament it received broad political support; all parties except for the Progress Party voted in favour of the ruling.

23 The motion was to expand the functioning sphere of Article 21 in the Gender Equality Act to include also all company boards, however only requiring at least 25% of each gender, and not 40% which was the case for publicly appointed boards, etc. (White paper from the Ministry of Children and Family Affairs, 1999) 24 A government coalition composed of the three parties in the middle of the party spectrum in Norwegian politics, the Center Party, The Liberal Party and the Christian Democratic Party. 25 White paper from the Ministry of Children and Family Affairs 2001: Gender representation in public limited companies, state limited companies, and state businesses, etc., proposal to change the Company’s Act and some other acts. (Høring: Kjønnsrepresentasjon i styret i allmennaksjeselskaper, statsaksjeselskaper og statsforetak, m.v. – forslag til endringer i allmennaskjesloven og i enkelte andre lover. (BFD, 2001). 26 A Conservative-Centre government coalition composed of the Conservative Party, the Liberal Party and the Christian Democratic Party. 27 Proposition to parliament, Nr. 97 (2002-2003). Government proposition on the revision of the Company’s Act, Ministry of Trade and Industry. (Ot. prp. nr 97 (2002-2003). Om lov om endringer i lov 13. juni 1997 nr. 44 om aksjeselskaper, lov 13. juni 1997 nr. 45 om allmennaksjeselskaper og i enkelte andre lover (likestilling i styrer i statsaksjeselskaper, statsforetak, allmennaksjeselskaper mv.). Oslo: Næringsdepartementet.)

110 

 The Making of Gender Quotas for Corporate Boards in Norway

Paying attention to the temporal context is about sequences and timing, as well as about political actors’ ability to seize the moment. In the policy process that led to the adoption of gender quotas for corporate boards, political agency was of crucial importance at two particular points in time. First, two politicians from the Christian Democratic Party, replacing each other as ministers of gender equality, played significant roles in the preparation of the gender quota legislation: Valgerd Svarstad Haugland and Laila Dåvøy. In 1999 the Minister of Children and Family Affairs, Valgerd Svarstad Haugland (1997-2000), from the Christian Democratic Party, proposed gender quotas in Norwegian corporate boards to be included in the Gender Equality Act. This proposition was part of the preparation for the major revision of the Gender Equality Act. The most obvious reason for why Minister Haugland proposed such a radical and controversial reform was the fact that gender equality was part of her portfolio as Minister of Children and Family Affairs. The issue became salient in connection with the process leading up to the revision of the Gender Equality Act. The first record where gender quotas on corporate boards was suggested as a legal possibility is found in an initial letter from the Gender Equality Ombud in connection with the preparation of the revision of the Gender Equality Act. The Gender Equality Ombud, Anne Lise Ryel, and the Director of the Centre for Gender Equality, Ingunn Yssen, were central in launching the possibility of legally regulating the gender composition of corporate boards, in the first place (Sørensen 2011). Women and management were, as mentioned earlier a hot public issue at that time. Actually one of the few central gender equality issues of the Christian Democratic Party, a party generally more concerned with traditional family values than with gender equality, is women in management (Vik 2012). Another concern relates to Minister Haugland’s more personal motives at that particular point in time. In many ways she was badly in need of a powerful gender equality issue to balance her reputation at that time of being anti-gender equality. Her “bad” reputation was due to her central position and responsibility for introducing the cash-for-care reform, a core issue of the Christian Democratic Party, representing their emphasis on traditional family values.28 There was an intense and heated debate surrounding the cash-for-care initiative, which came into force in 1998. A main argument in the public debate was that the arrangement meant a backlash for the Norwegian gender equality project (Ellingsæter, 2006, and this volume). An interpretation of Minister Haugland’s innovative endeavours for introducing gender quotas for corporate boards was that she was in need of strengthen and confirm her authority in relation to gender equality policies. After the possible regulation of gender balance in corporate boards was launched it was strongly supported by important stake-holders in ways that made it appear as though the political process proceeded on in its own dynamics. In between the

28 The cash benefit reform provided all families with children between 1 and 3 years with a monthly allowance, similar to the cost of state subsidies per child in state subsidized day care.



Discussion  

 111

initial phase and the final adoption, the legislation was prepared in silence. The legal aspects were assessed. The Labour Party continued this process and left its mark on it by changing the target from at least 25 to 40 per cent representation of each gender. After a legal assessment by the Ministry of Justice, the scope of the legislation was delimited to include only publicly owned and public limited (mostly listed) companies. The second and final phase has been hailed as paramount in the international press,29 especially the final acts and moves of the Minister of Trade and Industry, Ansgar Gabrielsen, from the Conservative Party. Shortly before the law proposition was to be discussed by the government, Gabrielsen gave an interview to the (then) largest Norwegian newspaper (VG 22.2.2002), where he uttered that he was “sick and tired of the male dominance in business life”.30 This was an unexpected and vigorous initiative from a male politician from the Conservative Party, who enjoyed high esteem in business circles. The uncertainty regarding whether the gender quota act would finally be passed in Parliament, rested on the rather heavy opposition that was expressed against the gender quota law proposition, especially from within economic life. The Conservative Party has a strong tradition of being attentive to the opinions within the business sector, although there were internal party controversies on the gender quota issue. Gabrielsen’s sudden initiative in favour of the quota act in the final phase of the political process, and his claim of the problems of male dominance in Norwegian economic life, probably were decisive in the final government round. In the final phase of the political process, this move probably managed to tip a divided government in favour of the quota legislation (see Magma 2010). For the second time the initiative was pushed forward from where it was not expected. Finally only the members of Parliament of the Progress Party voted against it, although some of the Conservative Party’s MPs probably felt obligated by the conclusion of Minister Gabrielsen and the government. Nonetheless, a long political process was in motion before Minister Gabrielsen entered the stage. The importance of his role for the final result, the adoption of mandatory gender quotas on Norwegian corporate boards tends to be exaggerated. Still, no one really knows what would have happened if it were not for his manoeuvers in the final phase of the political process.

6.4 Discussion The time and the timing of events were crucial for the making of gender quotas for corporate boards in Norway. For the policy process to be established, three critical moments that more or less coincided in time, have been identified as important. First,

29 See i.e. The Economist March 13, 2010, Time Magazine, April 26, 2010, Der Spiegel, July 8, 2010 30 http://www.vg.no/nyheter/innenriks/artikkel.php?artid=3024189

112 

 The Making of Gender Quotas for Corporate Boards in Norway

the highly profiled and heated media debate on women in management throughout the 1990s constituted an important background for why policy-makers started to search for policy initiatives to promote gender balance in economic decision-making positions. Particularly the discrepancy between the country’s self-image of being particularly successful in affairs of gender equality and the persistent male dominance in top positions in the economy, spawned political awareness about this as a problem and the need to promote policy innovations. The strong national tradition for applying gender quota measures directed attention in that particular direction. Second the preparation for the revision of the Gender Equality Act constituted a departure point for innovative thinking about the possibilities and limitations of gender equality legislation. In this initial phase central actors within the gender equality machinery (Ryel [Gender Equality Ombud] and Yssen [Director of Centre for Gender Equality]) played important roles as participants in the preparation of revision of the Gender Equality Act and as innovators. Thirdly, the deregulation of the economy, fuelled worries about the gender effects of neoliberal policies, and brought to the fore policies to counteract the limited scope of article 21 in the Gender Equality Act. In sum, the women in management debate, the revision of the Gender Equality Act and the effects of deregulation were important for preparing the ground for a novel policy initiative. Nonetheless, the adoption of gender quotas for corporate boards can be analyzed simultaneously as a continuation and a break with the Norwegian gender equality policy tradition. On the one hand, gender quotas for corporate boards builds on a policy legacy of adopting gender quotas. On the other hand, the expansion of such regulations beyond public boards and commissions can be viewed as a break with the institutionalization of boundaries between the state and the business sector. Moreover, it may appear counter to intuition that the scope of gender quota legislation was enlarged in a period otherwise characterized by deregulation and a weakening of political regulations. However, a main argument for deregulation is to provide similar conditions for companies irrespective of type and structure of ownership. Thus in the Norwegian context, deregulation (without gender quotas for corporate boards) would imply a delimiting scope of the legislation that regulates the gender composition of corporate boards. The extensive state ownership in Norway emphasizes the impact deregulation otherwise would have had on delimiting the scope of gender equality policy. The alternative, which would be to institute particular rules for publicly owned companies, would imply (at least in theory) a twist in the conditions for competition and were therefore not considered a realistic alternative. Hence, to avoid diminishing the impact sphere of gender equality policy, expansion of the gender quota policy emerged as an alternative solution. Thus, mandatory gender quotas on Norwegian corporate boards came forth as a national regulative response to the possible unintended gender equality consequences of the processes of deregulation. As argued by Vogel (1996, p. 3), deregulation often combines

Conclusion 

 113

liberalization with re-regulations, “Hence we have wound up with freer markets and more rules”. Exactly because the adoption of gender quotas for corporate boards was at odds with the Norwegian system of industrial relations, it was for some time uncertain what the final result would be of the policy process. A final concern is with the role of political agency in this particular policy process. Significant reforms are often made possible through deviance from the ideological main path, often with the consequence that large reforms and new directions of policies are carried through in a smoother manner than otherwise would have been the case. The ways in which powerful leaders can prepare the ground for significant changes due to sudden and vigorous initiatives are sometimes referred to as the “Nixon goes to China” dynamic.31 In this case, a progressive gender equality reform was proposed by a Centre-Conservative government coalition. This constituted a breach with an established “division of labour” within Norwegian politics, where the policies of gender equality had been dominated by the left-wing parties, while the interests, rights and self-regulation of business and private owners has been the concern of the right-wing parties. Thus, it was sensational that a bold gender quota initiative came from a minister from the Christian Democratic Party. Many of the political actors who are normally sceptical of regulations and gender equality initiatives may have become somewhat confused and more hesitant than they otherwise would have been if the law reform had been issued by the Labour Party, generally acknowledged for being proregulation and pro-gender quotas. This bewildering situation was continued by the second unexpected move in this political process, staged by Minister Gabrielsen, with a strong reputation for being a solid, conservative Minister of Trade and Industry and without any prior public engagement in matters of gender equality. This contributed to create a basis for the broad majority behind the mandatory gender quota on corporate board proposition.

6.5 Conclusion This chapter has described and explored the interplay of the contextual factors and processes that were decisive in the policy process that led to the making of gender quotas for corporate boards in Norway. This reform has been significant in the national context, by the way the gender equality institution managed to expand in

31 “Nixon goes to China” refers to incidents where significant changes occur as the result of the enhanced legitimacy possessed by powerful leaders, as well as the windows of opportunity which may open when politicians foster unexpected initiatives. This is what happened in 1972 when President Nixon went to China and met with Chairman Mao, illustrating a breakthrough for a peace process in the relationship between the two countries.

114 

 The Making of Gender Quotas for Corporate Boards in Norway

an era otherwise characterized by relative weakening of political governance and increased liberalization of the economy. To some extent this tale emphasizes the strength of the gender equality institution in Norway today. Although the political process was clearly featured by uncertainty from the moment it entered the policy scene until adoption and implementation, arguments on gender equality triumphed arguments concerned with the autonomy of the business sector with the result that mandatory gender quotas for corporate boards were adopted. A final question would be whether not a radical measure, such as gender quotas for corporate boards, is a sign of a significant drop in the legitimacy of male dominance – a drop in legitimacy following decades of gender equality progress. The enormous international attention on gender quotas for corporate boards, as witnessed through the spread across Europe of gender quotas for corporate boards, is perhaps telling about a new turn in the gender equality debate – and of its’ status in general.

References 

 115

References Ahern, K. R., & Dittmar, A. K. (2012). The Changing of the Boards: The value effect of massive exogenous shock. Quarterly Journal of Economics, 127(1), 137-197. Armstrong, J., & Walby, S. (2012). Gender Quotas in Management Boards. Note. Directorate-General for Internal Policies, Policy Department, Citizen’s Rights and Constitutional Affairs. Retrieved from http://www.europarl.europa.eu/document/activities/cont/201202/20120216ATT38420/2 0120216ATT38420EN.pdf Birkelund, G. E., & Sandnes, T. (2003). Paradoxes of Welfare States and Equal Opportunities: Gender and Managerial Power in Norway and the USA. In Comparative Social Research, 21, 203- 242. Cvijanovic, A. (2009). Rettferdig og rimelig? – om kjønnskvotering i styrene i allmennaksjeselskap. Masteroppgave i politikk og samfunnsendring, Fakultet for samfunnsvitenskap, Høgskolen i Bodø. Dahlerup, D., & Freindevall, L. (2005). Quotes as a Fast Track to Equal Representation for Women. Why Scandinavia is no Longer the Model, International Feminist Journal of Politics, 7(1), 1-22. Dale-Olsen, H., Schøne, P., & Verner, M. (2013). «Diversity among Norwegian Boards of Directors: Does a Quota for Women Improve Firm Performance?” Feminist Economics, doi:10.1080/135457 01.2013.830188 (2013). Derthick, M., & Quirk, P. J. (1985). The Politics of Deregulation. Washington D.C.: The Brookings Institution. Ellingsæter, A. L. (2006). The Norwegian child care regime and its paradoxes. In A. L. Ellingsæter & A. Leira (Eds.), 2006. Politicising Parenthood in Scandinavia. Bristol, UK: Policy Press. Ekberg, E., Gulbrandsen, T., & Vatnaland, J. (2003). Næringslivet mellom marked og politikk. Oslo: Gyldendal forlag. Engelstad, F. (2011). Kan kjønnskvotering i næringslivet forsvares? En normative analyse. Tidskrift for kjønnsforskning, 35(2), 120-140. Engelstad, F. (2012). Limits to state intervention into the private sector economy: Aspects of property rights in social democratic societies. In F. Engelstad & M. Teigen (Eds.) Firms, Boards and Gender Quotas: Comparative Perspectives. Comparative Social Research, Bingley: Emerald. Evenrud, M. (2010). Politisk diskurs forut for innføringen av lov om representasjon av begge kjønn I styrene til allmennaksjeselskap. Masteroppgave. Oslo: University of Oslo. EWL report. (2012). Women on Boards in Europe from a snails’ pace to a giant leap? EWL report on Progress, Gaps and Good Practice. European Women’s Lobby February 2012. Fagan, C., Menéndez, M. C. G., & Ansón, S. G. (2012). Women on Corporate Boards and in Top Management. European Trends and Policy. London: Palgrave. Franceschet, S., & Piscopo, J. (2013). Equality, Democracy and the Broadening and Deepening of Gender Quotas. Politics & Gender, 9(3), 310-316. Hernes, H. (1987). Welfare State and Woman Power. Essays in State Feminism. Oslo: Norwegian University Press. Jenson, J. (2010). Diffusing Ideas for After Neoliberalism. The Social Investment perspective in Europe and Latin America. Global Social Policy, 10(1), 59-84. Krook, M. L. (2008). Quota Laws for Women in Politics: Implications for Female Practice. Social Politics, 15(3), 345-368. Krook, M. L. (2009). Quotas for Women in Politics: Gender and Candidate Selection Reform Worldwide. New York: Oxford University Press. Krizsan, A., Skjeie, H., & Squires, J. (2012). Institutionalizing Intersectionality. The Changing Nature of European Equality Regimes. Basingstoke: Palgrave Macmillan.

116 

 References

Magma. (2010). Da kvoteringsloven kom til verden: En politisk styrtfødsel. Intervju med Ansgar Gabrielsen i tidsskriftet Magma, nr. 7, 2010, 15-19. (Interview with Ansgar Gabrielsen in the scientific management journal, Magma). Matsa, D., & Miller, A. R. (2013). “A Female Style in Corporate Leadership? Evidence from Quotas”, American Economic Journal of Applied Economics, 5(3), 136-169. Mahoney, J., & Thelen, K. (2010). A Theory of Gradual Institutional Change. In K. Thelen & J. Mahoney. Explaining Institutional Change. Ambiguity, Agency and Power. Cambridge: Cambridge University Press, 1-37. McBride, D., & Mazur, A. G. (2010). The Politics of State Feminism. Philadelphia: Temple University Press. Niskanen, K. (Ed.). (2011). Gender and Power in the Nordic Countries – with focus on politics and business. Oslo: NIKK-publication 2011: 1. http://www.nikk.no/filestore/Publikasjoner/NIKKPublications_GenderandpowerintheNordiccountries.pdf Pierson, P. (2000). Path-dependence, Increasing Returns and the Study of Politics. American Political Science Review, 14(1), 73-93. Pierson, P. (2004). Politics in time. History, institutions, and social analysis. Princeton and Oxford: Princeton University Press. Schmidt, V. (2008). Discursive Institutionalism: The Explanatory Power of Ideas and Discourse. Annual Review of Political Science, 11, 303-326. Skjeie, H., & Teigen, M. (2003). Menn i mellom. Mannsdominans og likestillingspolitikk. Oslo: Gyldendal Akademisk. Simon-Kumar, R. (2011). The Analytics of “Gendering” the Post-Neoliberal State. Social Politics, 18(3), 441-468. Stjernø, S. (2005). Solidarity in Europe: The History of an Idea. Cambridge, UK: Cambridge University Press. Sørensen, S. Ø. (2011). Statsfeminismens møte med næringslivet. Bakgrunnen og gjennombruddet for kjønnskvotering i bedriftsstyrer som politisk reform. Tidsskrift for kjønnsforskning, 35(2), 102-119. Teigen, M. (2002). Kvotering til styreverv – Mellom offentlig og privat handlefrihet. Tidsskrift for samfunnsforskning, 43(1), 73-104. Teigen, M., & Wängnerud, L. (2009). Tracing Gender Equality Cultures: Elite Perceptions of Gender Equality in Norway and Sweden, Politics & Gender, 5, 21-44. Teigen, M. (2011). Gender Quotas on Corporate Boards. In K. Niskanen (Ed.), 2011. Gender and Power in the Nordic Countries. With focus on politics and business. Oslo: NIKK-publication. http://www.nikk.no/filestore/Publikasjoner/NIKKPublications_GenderandpowerintheNordiccountries.pdf Teigen, M. (2012). Gender Quotas for Corporate Boards in Norway – Innovative Gender Equality reform. In C. Fagan, M. C. G. Menéndez & G. S. Ansón. (Eds.) Women on Corporate Boards and in Top Management: European Trends and Policy. London: Palgrave. Teigen, M. (2012). Gender Quotas in Corporate Boards – On the diffusion of a distinct national policy reform. In F. Engelstad & M. Teigen. (Eds.) Firms, Boards and Gender Quotas: Comparative Perspectives. Comparative Social Research, Bingley: Emerald. Terjesen, S., Aguilera, R.V., & Lorenz, R. (2014). Legislating a Woman’s Seat on the Board: Institutional Factors Driving Gender Quotas for Board of Directors. To be published in Journal of Business Ethics. Thelen, K. (1999). Historical Institutionalism in Comparative Politics. Annual Review of Political Science, 2, 369-404. Thelen, K., & Mahoney, J. (2010). Explaining Institutional Change. Ambiguity, Agency and Power. Cambridge: Cambridge University Press.

References 

 117

Thelen, K. (2014). Varieties of liberalization: The New Politics of Social Solidarity. New York: Cambridge University Press. Vik, M. L. (2012). Kjønnslikhet, kvinneulikhet, og mangfold – «velment og forent» eller på tvers og tverke? En komparativ studie av norsk likestillingspolitikk 1981-2013. Masteroppgave, Institutt for statsvitenskap, Universitetet i Oslo. Vogel, S. K. (1996). Freer Markets, More Rules. Regulatory Reform in Advanced Industrial Countries. Ithaka, NY: Cornell University Press. Waylen, G. (2014). Informal Institutions, Institutional Change, and Gender Equality. Political Research Quarterly, 67, 212-223.

Cathrine Holst

7 Institutional Variation and Normative Theory: Lessons from a Local Equal Pay Controversy Equal pay for equal work, or work of equal value, is a central slogan for the women’s movement and an established principle in international law.1 How can equal pay be justified? Which normative concerns and arguments speak against the equal pay principle; what speaks in favour of it? In the international academic literature normative discussions of equal pay, also referred to as “comparable worth”, go on in different branches of gender studies, in economics, as well as in moral and political philosophy. A central ambition of this chapter is to investigate the relationship between these debates going on in the academic field and the equal pay controversy as it has been unfolding in Norway. A closer comparison shows that familiar arguments in academic exchanges are present in this local controversy but that the Norwegian equal pay exchanges are characterized by a richer normative structure in that a broader set of reasonable concerns and arguments are raised. How can this discrepancy be understood? Why is the Norwegian controversy different and seemingly more advanced? Several hypotheses are discussed, and it is argued that the different grammar of the academic and the “local” debate in this case is hard to explain without reference to institutional variation and the relationship between institutional and normative structure: The added concerns and arguments of the Norwegian equal pay debate have as their template a particular institutionalization of working life relations and wage setting. This finding has general implications for how we conceive of the relationship between empirical investigations and normative argumentation and theorizing. More specifically, it is argued for a particular relationship between awareness of and inquiries into institutional variation and the strongest possible testing of normative arguments and principles.

1 The principle was included in the constitution of the International Labor Organization (ILO) in 1919. Convention Number 100 concerning equal remuneration for men and women workers for work of equal value (ILO-100) came into force in 1952. In the Universal Declaration of Human Rights from 1948 the principle was made part of article 23 (“Everyone, without any discrimination, has the right to equal pay for equal work”), whereas article 7 of the International Covenant on Economic, Social and Cultural Rights from 1966 guarantees “fair wages and equal remuneration for work of equal value”, and the Convention on the Elimination of All Forms of Discrimination against Women from 1979, article 11, ensures the right of women and men to “equal remuneration, including benefits, and to equal treatment in respect of work of equal value”. Other equal pay articles in international law includes the famous Article 119 on equal pay (later Article 141 EC, now Article 157 TFEU) in the Treaties of Rome (1957), the only article on social policy in the European Union’s (EU) original treaty placing a direct obligation on the member states. © 2015 Cathrine Holst This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Is Equal Pay Justified? The Standard Academic Debate on Comparable Worth 

 119

The first part of this chapter presents the central arguments for and against equal pay or comparable worth in international academic literature. The second part elaborates on the Norwegian equal pay controversy. What are the key arguments? And how do these arguments differ from the standard pro and con equal pay arguments that we know from academic discourse? The third part discusses how the identified discrepancy between the academic and “the local” can be explained. Particular weight is put on the relationship between the characteristics of the Norwegian comparable worth controversy and the Norwegian working life institutionalization, and how this relationship – in this case – contributes to a periphery advantage. The final part revisits the debate on the relationship between normative theory and empirical scholarship and takes up the role of institutional analysis.

7.1 Is Equal Pay Justified? The Standard Academic Debate on Comparable Worth Wage structure and income patterns are important research topics in empirical social science, not least among economists and sociologists. Here “equal pay” occurs, when it occurs, as a descriptive category used to describe this or that wage distribution according to some more detailed operational definition of “equal pay”, or in explanations, typically in studies that seek to explain why instances of “unequal pay” occur. The focus here is rather on the normative debate of whether “equal pay” or “comparable worth” is worth defending. Contributions to this debate come from different branches of gender studies and feminist theory, including feminist economics, but also from other branches of economics and from ethics and political theory. The ambition of this section is not to make a detailed review of the available literature, and even less to provide a normative assessment of whether equal pay is defensible. Rather, the idea is to lay out the more or less standard international academic debate on why – or why not – comparable worth is – or is not – worth defending. What are the arguments on both sides that are typically found? How does the equal pay controversy conventionally play itself out in academic discourse?

7.1.1 Standard Contra Arguments Generally, three standard arguments can be listed on the sceptical side: a justice argument, a theoretical argument and an efficiency argument. Some critics subscribe to all three arguments or at least more than one (for example, Aaron & Lougy, 1986; Hill & Killingworth, 1989; Paul, 1993), while some emphasize only one of them, for example the efficiency argument, while explicitly dismissing one or more of the

120 

 Institutional Variation and Normative Theory: Lessons from a Local Equal Pay Controversy

others (for example, Roback, 1986; Jacobsen, 1998). The justice argument or line of reasoning dismisses equal pay on the basis of a certain conception of justice and moral right. The basic contention is that well-functioning markets distribute pay and other goods fairly, and thus that work should be recognized and paid according to its market value and not according to market-independent standards of value and equal value (Paul, 1993, pp. 9-22; Aldrich & Buchele, 1986; Killingworth, 1990; Pincus & Shaw, 1998; O’Neill, 2000). This gives little scope or no place for equal pay ideas or policies. At the heart of this argument is a defence of the primary moral status of private autonomy and individual liberties, including economic liberties, and the assumption that, generally speaking, the market value of work is the “right” value, since the market value of transaction objects, be it goods, services, capital or labour, are the aggregated outcomes of individuals applying their rightful liberties. From this perspective any equal pay project is suspect, opening up for immoral and disrespectful private sphere intrusions. This approach, it must be emphasized, does not allow for direct discrimination: If women are paid less because of sexist prejudices among employers, this is typically considered a problem. On this basis it supports anti-discrimination laws. In addition there is a general belief that the free market mechanism over time will work against prejudice-based discrimination: Assuming that employers primarily are after maximal productivity and profits, the belief is that employers over time will adjust what may be prejudiced-based wage setting to what is optimal for productivity. If so, what may be of aggregated wage differences between women and men represents no injustice. Such differences would then be the result of systematic differences in work-related preferences among male and female employees, women opting for lower positions in low productivity industries and sectors, or crowding in some industries or sectors, increasing labour supply and creating downward pressure on wages in these parts of the economy. There is also the possibility that women are actually less productive, due to, for example, a stronger family orientation. Secondly, there is the theoretical argument that there are no objective criteria of value that different categories of work can be assessed as equal or unequal with reference to (Paul, 1993, pp. 23-27; Aldrich & Buchele, 1986; Killingworth, 1990; Pincus & Shaw, 1998; O’Neill, 2000; but see also England, 1992). Any proposals of evaluation criteria will be controversial and partial, in accordance with some citizens’ subjective preferences, and at odds with others: There are no clear-cut ways of deciding beyond disagreement the value of different work characteristics, tasks and responsibilities or of different kinds of training and education. Some may value industrial work above nursing or accounting, teaching above engineering or banking, leadership qualities and practical experience above academic training, while for others it is the other way around. A secondary justice argument against equal pay follows from this: If a state is to enforce equal pay based on criteria of value and equal value as something different from market value, it cannot do so impartially. A state enforcing equal pay has thus stopped short of being the neutral arbitrator a just state ought to be and turned into



Is Equal Pay Justified? The Standard Academic Debate on Comparable Worth 

 121

a technocratic regulator operating on the basis of controversial notions of what is valuable and good. Finally, there is the efficiency or productivity argument that equal pay will result in intolerable productivity losses and inefficiencies (see for example Aldrich & Buchele, 1986; Wilborn, 1986; Hill & Killingworth, 1989; Killingworth, 1990; O’Neill, 2000; but also Sorensen, 1994; Jacobsen, 1998). The contention is that relatively unregulated markets maximize productivity – that is, if such markets work as they are supposed to, labour moves from low productive firms and industries to more highly paid positions in high productive firms and industries. Societies characterized by too many or the wrong kind of legal and administrative restrictions and regulations on free labour movement or by undue union influence on wage structure are thus vulnerable to loss in productivity, since wage setting on the basis of market-external political concerns increases the likelihood of over-payment of unproductive work and under-payment of productive work. Within such a framing, equal pay policies are regarded as highly problematic. In the short run they would simply result in lower productivity. In the longer run they would, as far as they are effective, be detrimental to women’s interests, since they contribute to increasing the price on female labour and thus create incentives for employers to cut down and re-invest in ways that make them less dependent on an “overpriced” workforce. Equal pay sceptics often add to such general arguments analyses or references to analyses of concrete equal pay experiments where the long-term effects on women’s payment are uncertain, zero or negative (for example Killingworth, 1990; Rhoads, 1993; Aaron & Lougy, 1986; but also Hammer, 1986; Gunderson, 1994; O’Neill, 2000). It must be stressed that this line of reasoning – equal pay scepticism based on efficiency concerns – can come together with the views on discrimination and free choice typical among proponents of the justice argument against equal pay, but also with very different “progressive” and even explicit feminist views (for example Milkman, 1990; Jacobsen, 1998; Heath, 2009). Feminist equal pay sceptics figure there to be both direct discrimination and prejudiced-based discrimination and other social mechanisms contributing to unequal opportunities for women and men in economy and society. However, as they see it, comparable worth is not the adequate remedy, and they support instead equal opportunities policies, including affirmative action (see also Lapidus & Figart, 1998; Jacobsen, 1998).

7.1.2 Standard Pro Arguments Equal pay proponents’ standard reply to equal pay sceptics can also be summed up in a package of three arguments. A justice argument is central as well for those in favour of equal pay; in addition comes an empirical and a pragmatic argument. As for the justice argument, it is claimed, basically, that equal recognition of, and so equal pay for, work that is similar or of equal value, irrespective of its market value,

122 

 Institutional Variation and Normative Theory: Lessons from a Local Equal Pay Controversy

is a basic social justice concern (see Fraser, 2003; Krebs, 2003; Honneth, 2003 for different versions of this argument, but also Treiman & Hartmann, 1981; Hartmann, 1985; Weinzweig, 1987; England, 1992; Pincus & Shaw, 1998; Macdonald & Merrill, 2002).2 Underlying here is a norm of equal treatment, that equal cases ought to be treated equally (see van der Vleuten, 2008), implying also, or so it is assumed or argued, that work that is “equal” (similar or of equal value) is to be treated “equally” in the sense that it is to be given equal recognition. The second step is to establish equal pay as a key indicator of equal recognition. Work is recognized as being equal or of equal value by means of being paid equally, and the other way around: When work is paid unequally, this is an expression of unequal recognition. Finally, it is pointed out how unregulated markets in contemporary societies produce distributive patterns that reflect the fact that in these societies, work that is similar or of equal value is frequently paid unequally. There is thus nothing inherently just with these patterns, equal pay proponents would say. Rather, if we care about equal pay, along with a set of other just concerns, market-correcting measures are not only permissible but also required. The second, more empirically oriented argument in favour of comparable worth, often pursued together with justice considerations, is focusing further on what proponents regard as sceptics’ misconceived ideas of the role of markets. Here the target is what critics believe are idealized, empirically ill-founded assumptions about the actual working and institutionalization of markets and market behaviour (see, for example, Johansen, 1984; Wilborn, 1986; Acker, 1989; Evans & Nelson, 1989; England, 1992; Blum, 1993; Mutari, Figart & Power, 2001; Karamessini & Ioakimoglou, 2007). This pertains both to the idea that free markets over time will work against prejudices and so result in “unbiased” employment practices regulated purely by productivity concerns and to the idea that gender segregation in labour markets, including the crowding of women in lower positions in low productivity industries and sectors, therefore will be the result of relatively autonomous individual choices among workers. Equal pay proponents refer here, on the demand side, to persistent prejudices and biases against women and other groups in the labour market, and, on the supply side, to how women’s working life adaptations reflect unequal opportunities for women and men and not simply their “free choice”, even in the absence of direct discrimination. Due to this, they would say, the contention that labour is moving “freely” and regularly to high productivity firms, industries and sectors in absence of equal pay and other “distorting” market regulations, is somewhat of a myth. In addition, it is well known how labour markets even in free market economies are regulated by legislation and structured by professional and union bargaining, not

2 Along with other such concerns. For example Nancy Fraser (2003) argues for justice understood as “participatory parity”, and for “redistribution”, “recognition” and “democratization” as preconditions for realizing justice. Comparable worth figures here as a “recognition” claim.



A Local Controversy: Equal Pay in a Scandinavian Context 

 123

least in the public sector (for example, Milkman, 1990). Markets and market behaviour are thus already “distorted”, and regulating measures such as equal pay legislation must be evaluated on the basis of their applicability and merits in real-life markets, not on what we could expect in idealized situations. Hence, the approach to markets among equal pay proponents and “progressive” equal pay sceptics do not necessarily differ that much; what divides them is the more specific assessment of comparable worth. Finally, when all the above has been said, a pragmatic argument is typically added by proponents, meant to target sceptics’ theoretical argument that there are no impartial, objective criteria of value that different categories of work can be assessed as equal or unequal with reference to. Or rather, even if there are philosophical contributions with ambitions of connecting comparable worth to non-partial, intersubjective claims of equal recognition and transcultural standards of ethical life (see Honneth, 2003), the most common answer to the theoretical there-are-no-objective-criteria argument is that this argument is exactly theoretical, “philosophical” or “academic”. In practice agreements on practical, operational indicators can be and have been reached (see, for example, Hartmann, 1985; England, 1992; Sorensen, 1994). Equal pay proponents can here refer to how job evaluation schemes have been developed and included in national legislation and U.S. state legislation, as well as at the local level in union-firm bargaining, and how this seems to contribute to reducing gender-based wage gaps (Schwab, 1985; Wilborn, 1986; Michael, Hartmann & O’Farrell, 1989; Rhoads, 1993; Gunderson, 1994; Lapidus & Figart, 1998). Moreover, comparable worth seems in the end to reflect some common moral intuitions, irrespective of whether philosophers manage to grapple with them argumentatively or not, reflected not least in how the equal pay principle is firmly established in international law.

7.2 A Local Controversy: Equal Pay in a Scandinavian Context We move now to the equal pay controversy in Norway. Norway ratified Convention 100 of the International Labour Organization (ILO-100) in 1959, and § 5 of the Norwegian Gender Equality Act passed in 1978 stated accordingly that “women and men shall have equal pay3 for the same work or work of equal value”. How the expression “work of equal value” was to be understood more specifically was not spelled out in the article, but in preparatory documents “work of equal value” was defined as work that appeared as fairly “similar”, and it was argued that the value of work could not

3 “Pay” refers here to ”ordinary remuneration for work as well as all other supplements or advantages or other benefits provided by the employer”.

124 

 Institutional Variation and Normative Theory: Lessons from a Local Equal Pay Controversy

be compared meaningfully across trades and professions.4 Section 5 was however amended in 2002 with reference to Norway’s international obligations and the EEA agreement,5 and the article now states specifically that “women and men in the same enterprise shall have equal pay for the same work or work of equal value” and that this “shall apply regardless of whether such work is connected with different trades or professions or whether the pay is regulated by different collective wage agreements”.6 The equal pay issue has been the topic of several rounds of rather extensive public deliberations in Norway. The question that interested me at the outset was the argumentative profile of these deliberations and how the standard arguments on equal pay as we know them from international academic exchanges fared in the Norwegian context. However, it soon became clear that a set of concerns and arguments absent from these exchanges were just as important or more important in the Norwegian debate. The data analyzed for the purpose of this chapter are in part existing studies and analyses of the 2002 amendment and the political process preceding it, including the formal public hearing and the parliamentary debate (Tenden, 2001; Høgsnes, 2002; Skjeie & Teigen, 2003) and in part two public reports on equal pay, one from 1997 on job evaluation schemes and the 2008 Equal Pay Commission7 and the public hearing of the latter. As for the more particular proponents of different views on equal pay in the Norwegian case, they will be introduced only occasionally and in passing: The point here is not to elaborate on the details of the political process, the proposed policies or the more specific role of different actors but to expose some major concerns and arguments on equal pay that are raised in this local controversy and that are seemingly uncovered by the three-plus-three list in the previous section.

7.2.1 Additional Contra Concerns and Arguments If we start with the equal pay scepticism, all the three standard critical arguments are also raised in the Norwegian case. The efficiency or productivity argument against equal pay regulations is the most common of the three; it is central to equal pay

4 Ot.prp. nr. 1 (1977-78) Lov om likestilling mellom kjønnene. 5 The Agreement of the European Economic Area from 1994 brings together the EU member states and the EEA states that are members of EFTA (the European Free Trade Association) in the Internal Market. 6 Ot.prp. nr. 6 (2001-2002) Om endringer i likestillingsloven mv. 7 NOU 1997: 10 Arbeidsvurdering som virkemiddel for likelønn (Job evaluation as equal pay policy) and NOU 2008: 6 Kjønn og lønn. Fakta, analyser og virkemidler for likelønn (Gender and wages. Facts, analyses and policies for equal pay). The commissions appointed to produce these reports consisted of social scientists, social partners and civil society representatives, central administration bureaucrats and experienced ex-politicians.



A Local Controversy: Equal Pay in a Scandinavian Context 

 125

critics among the social partners, state agencies and ministries, and civil society organizations, but is even highlighted by equal pay proponents as a worry and “second thought” – for example, in both the equal pay-friendly public reports.8 The standard theoretical argument, that there are no objective criteria of job value, and the standard moral or justice argument about the right to economic freedom and equal pay as an illegitimate private sphere intrusion, are also mobilized, even if less frequently. The justice argument constitutes the backbone of employer organizations’ equal pay opposition9 and is also raised elsewhere – for example, in the sceptical equal pay minority view of the Equal Pay Commission, whereas the theoretical argument is mentioned more seldom and most often in passing. However, this leaves out a central tenet of equal pay scepticism in the Norwegian controversy, which is the contention that equal pay policies come into conflict with the idea that pay should be set by means of collective struggles and negotiations between unions and employers. Equal pay sceptics point out how collective bargaining in a centralized system, sometimes with the state as a third negotiating partner, is key in “the Norwegian model” of wage setting and worry or believe that pushing equal pay legislation top-down will challenge or contribute to eroding a regime that has proven to be both effective and fair. That is, what is mobilized against equal pay within this framing is in part a moral or justice argument completely different from the standard justice argument; in part an instrumental argument of what a centralized collective bargaining system of wage setting is good for. The moral argument is conceptualizing free collective bargaining as a democratic right for workers: Central to democracy is the right to unionize, and workers’ organization and mobilization have earned unions their rightful power in society, including decisive influence in wage bargaining processes. This argument is typically pursued by some of the unions, including Norway’s largest union federation (LO),10 which came out on the sceptical side both in the process preceding the 2002 amendment and in the Equal Pay Commission hearing. These unions also emphasize what they regard as the progressive redistributive effects of the Norwegian wage setting system – free collective bargaining has contributed to reduced inequalities. Equal pay is thus framed to cause direct trouble for “democracy” – for societal democratization through organizing labour and power to unions – and, indirectly, also for “equality”, egalitarian social and economic distributions. In addition comparable worth redistributions are targeted directly as being substantively unjust as far as they, based on equations between job value and educational levels, imply raising wages in

8 An equal pay sceptic, President of BI Norwegian Business School Tom Colbjørnsen and member of the Equal Pay Commission, dissented from this Commission's equal pay-friendly conclusions. 9 Consider, for example, the hearing report of the Norwegian Confederation of Enterprise (NHO). 10 The Norwegian Confederation of Trade Unions.

126 

 Institutional Variation and Normative Theory: Lessons from a Local Equal Pay Controversy

female-dominated public sector educational groups relative to wage levels in manual work or in other groups. Finally, the bargaining wage setting model is claimed to increase both de facto legitimacy and overall national productivity. On these points, the equal pay sceptic unions are accompanied by other sceptics, such as the central employer organizations and public agencies and ministries external to the gender equality segment (the labour ministry, the finance ministry, etc.). Even proponents often include these concerns prominently in their discussions, even if they in the end conclude differently on the specific effects of equal pay regulations. The Norwegian collective bargaining system, its institutionalization and democratic and egalitarian underpinnings and implications, is presented as producing trust, political stability and a consensus culture that ensures swift and sound decision-making and high levels of citizens’ support. It is also presented as essential for the comparatively high productivity levels of the Norwegian economy. The latter is even stressed by employer organizations praising how “the Norwegian system” at its most optimal combines effective and long-term productivity serving collective bargaining procedures with flexible marketbased wage setting.

7.2.2 Additional Pro Concerns and Arguments Norwegian equal pay proponents also rely on the standard arguments familiar from international academic exchanges. Both in the equal pay-friendly public reports and in contributions to hearings by the gender equality ministry, the gender equality and anti-discrimination ombud, female-dominated public sector unions (nurses, teachers, etc.) and other equal pay proponents, we typically find the standard pro package of justice arguments and pragmatic and empirical arguments: Equal pay is a question of equal recognition and of what is right and just and a part of Norway’s international human rights obligations; equal pay regulations are more applicable than critics claim; and resistance against them is based on misleading ideas of market behaviour and how the market works. However, the richer set of concerns and arguments on the contra side requires and gets additional responses. Generally, equal pay proponents also start out with recognizing the merits of the Norwegian bargaining system and societal model and emphasize how comparable worth and equal pay policies should be thought of and implemented as far as possible within the normative and institutional framework of this model, and not in ways that radically depart from it. An expression of this consensus is how the principled equal recognition argument for comparable worth and equal pay as an international legal obligation play a secondary role in the two



A Local Controversy: Equal Pay in a Scandinavian Context 

 127

equal pay reports, in particular the 2008 Commission.11 The same pragmatic strategy is visible in both reports’ treatment of the standard theoretical argument against equal pay – that there are no objective criteria of job value. This argument is generally granted before both reports go on to discuss operational indicators, concrete experiments and equal pay regulation policy by policy. Another expression of the underlying consensus of the Norwegian model as a success model is also how the policy discussions not least of the Equal Pay Commission stand out as considerably less legalistic, less exclusively focused on comparable worth and less focused on the idea of targeting unequal pay by means of developing and applying job evaluation schemes than the policy discussions initiated by feminists and others in standard academic exchanges on equal pay. The focus is rather on how to implement equal pay through the bargaining system and with the state as a third partner (the 2008 Commission suggests, for example, that the state brings an “equal pay pot” to the negotiation table), in combination with a set of gender-sensitive family and labour market policies framed, not to ensure comparable worth directly, but to target the gender pay gap more generally. The use of job evaluation schemes by the social partners during wage negotiation is recommended by the 1997 Commission but toned down in the 2008 report and not included in its prioritized recommendations. Hence, when equal pay proponents go on targeting some of the redistributive effects and certain normative and institutional blind spots of the Norwegian collective bargaining model, they do so modestly and without challenging the model’s basic characteristics.12 First, it is pointed out how some groups, typically femaledominated public sector educational groups, have benefited less from this model. The central reason for this, critics argue, is the bargaining system’s inability to handle the distributive effects and unfairness caused by gender segregation in the labour market and patriarchal biases in wage setting processes, influencing employers and also male-dominated unions. Secondly, the democracy argument for the Norwegian bargaining system is problematized when it is pointed out how certain actors and certain (male) interests are systematically privileged as this system currently works, causing a gendered democratic deficit. Finally, equal pay proponents and sceptics agree on the productivity merits of the Norwegian bargaining system, but proponents see no reason why these merits will not persist if equal pay is implemented carefully and pragmatically, relying on established procedures, approaches and routines.

11 Compare here with hearing reports produced by some of the female-dominated public sector unions, feminist activist organizations and state feminist machineries, where these concerns are primary. 12 They seem thus to adapt a strategy that would typically result in “layering”, see Mahoney and Thelen (2009), Thelen (2009), and also Fredrik Engelstad’s contribution to this book.

128 

 Institutional Variation and Normative Theory: Lessons from a Local Equal Pay Controversy

7.3 Why is the Local Controversy Seemingly Richer? We have thus a situation where the Norwegian local pay controversy stands out not only as different from what we would expect on the basis of standard academic exchanges on what comparable worth is about, but seemingly also as richer or even more advanced: This local controversy does include the listed standard arguments but also a set of other arguments and concerns that seem to be important when deciding on what to think about the equal pay principle and equal pay policies. The question is how it turns out like this, and let me approach this question by anticipating and reflecting on some possible responses to the picture I have drawn in the two previous sections. “Are not international academic debates argumentatively superior to a messy national political controversy? You must have got it wrong”. Yes, contributions in international academic debates would no doubt and of course tend to hold higher academic standards than input produced by a civil society organization or a governmental agency in the sense that academic discourse arguments are typically (if not always), and also supposed to be, more stringently elaborated upon and more rigorously scrutinized and tested. This is also the case with the equal pay controversy if we compare, for example, philosophers’ accounts of the moral, individual, rightsbased case against equal pay with Norwegian business interests’ rough defence of free markets or the equal recognition argument for equal pay as conceptualized by Axel Honneth or Nancy Fraser with the attempts of formulating principle-based arguments among Norwegian equal pay activists. That the academic is superior to the local in this sense, even in the equal pay case, is not really put into question. The point is rather that the Norwegian controversy seems to introduce new concerns and arguments. On the contra-side, it introduces, as we have seen, a justice argument against equal pay legislation with reference to the right to free collective bargaining framed as a key democratic right for workers and workers’ unions. In addition to this comes a set of instrumental arguments for wage setting through collective bargaining processes (and not through equal pay regulations external to this process): a redistributive fairness argument, a de facto legitimacy argument and also a productivity argument that is very different from the standard productivity argument against equal pay. On the pro-side, the Norwegian controversy contributes primarily with broadening the scope of the pragmatic response (more and other measures than job evaluation schemes, equal pay policy making through bargaining processes, etc.) – equal pay, it is argued, is practically implementable in even more ways than what is standardly assumed, in addition to a pro-equal pay productivity argument. To be sure, the equal pay pro-camp often enough highlights how real life markets are already “perverted” and differ from idealized markets (i.e. the standard empirical pro-equal pay argument). However, what we have here in addition is a defence of such market “perversions”, or a certain combination of them, as “better” for economic sustainability, productivity and long-term profits



Why is the Local Controversy Seemingly Richer? 

 129

than “non-perverted” markets. In the Norwegian case many equal pay sceptics and proponents are indeed united in their defence of the Norwegian wage setting system (“the Norwegian model”) – despite standard views, they claim, the level of marketexternal intervention institutionalized by this system is compatible with and even stimulates productivity and growth. Where they differ then is in their assessment of the relationship between equal pay claims and this, as they see it, productivity enhancing system: Whereas equal pay sceptics worry that adaptations to such claims will erode the model, proponents believe that the model is robust and flexible enough to include them without compromising productivity. “But more advanced? You have not showed that the added arguments will tip the debate. You have not even shown that they are valid”. This is right. I have nowhere argued that this or that of the added argument in the end strengthens or weakens the case for equal pay. I have also not gone into assessing the more specific merits of each of the arguments. What is the more exact normative status of the right to free collective bargaining? Is it a proper “right”, and if so, what does it imply in the equal pay case? What is the causal relationship between centralized, tripartite bargaining on the one hand, and redistributive fairness, de facto legitimacy and productivity on the other? Would equal pay introduced as a state-sponsored equal pay pot during negotiations be effective? Would equal pay Norwegian-style regulations serve or hamper productivity? The added arguments raise these and other questions that have not been given any conclusive answers here, and the strength of the different arguments and the ultimate “tipping”, will of course depend decisively on these answers. With regard to some of the questions there will be research literature available to consider (for example on the role and effects of the Norwegian bargaining system), but other questions are highly open and left to be pursued properly. However, none of the added arguments stand out as obviously unreasonable or irrelevant – this is the (weaker) claim I want to make. Each of them may turn out to be “good” or not to hold water, but on the face of it they seem to deserve our attention and closer investigation. It is in this sense that the Norwegian controversy is considered “richer” and even “more advanced” than the standard academic debate in this chapter: It includes familiar arguments pro and contra equal pay, but also a set of additional concerns and arguments that are not clearly flawed or beside the point. “So the lesson then is that Scandinavian social democracy tends to produce normative deliberations that are richer?” No, no such general conclusions can be drawn. The argument here, and that will be spelled out in more detail below, is that normative deliberations and policy debates are typically embedded in local institutional contexts – institutional and normative structures are interconnected. However, the fact that this is so is at the outset equally compatible with narrowmindedness and provinciality – that is, to disregard or be ignorant of reasonable concerns and arguments that are or considered to be more “foreign” to local debates and contexts (as when liberal and multicultural critics claim, as they recurrently do, rightfully or not, that Norwegian public debate suffers from communitarian or

130 

 Institutional Variation and Normative Theory: Lessons from a Local Equal Pay Controversy

nationalistic biases). The point of this chapter’s discussions is that the Norwegian equal pay controversy specifically turned out to include concerns and arguments with a local institutional genesis in addition to the arguments we know from international exchanges, contributing to making the overall debate richer, but nothing is or can be claimed here about Norway, Scandinavia or social democracy implying that this will always be the case in these settings. “The contention then is that institutional and normative structures influence one another and that this can result in enriched normative deliberations under special circumstances”. Yes, this is exactly the point. Institutional theory and studies of variations in institutional regimes stress how institutional characteristics and constellations contribute to shaping discourse on norms and policies. A set of more particular studies give further reason to expect interconnections between working life and wage setting institutionalization and paths and angels in debates on distribution and redistribution of pay in particular (see Martin & Thelen, 2007; Thelen, 2009; Martin & Swank, 2010, 2012). The added arguments pro and con equal pay from the Norwegian controversy are illustrative of this: Both the proposed conflict between equal pay and democratic wage negotiations, productivity effects and implementation issues are discussed with the Norwegian tripartite centralized bargaining system as a central frame of reference (see Barth, Røed & Torp, 2002; Barth, Moene & Wallerstein, 2003; but also Martin & Swank, 2010). We have a situation, moreover, where standard equal pay arguments as we know them from academic discourse are added to these local concerns and approaches in the Norwegian controversy, but not the other way around: The added arguments from this local controversy are completely absent from the standard academic exchanges. There may be several reasons for this absence, but a hypothesis would be that it is related to the fact that Norway is “periphery” and Norwegian working life institutionalization is “untypical” and to the related fact that academic deliberations on the equal pay principle generally have non-Scandinavian and more “centre” wage setting institutions, regulations and procedures as their frame of reference, explicitly, or implicitly, not least the U.S. system. If so, this suggests an interpretation of the particular richness of the Norwegian equal pay controversy as the outcome of a periphery advantage – that is, as the more or less unintended effect of a situation where “centre” does what “centre” often does: namely, overlooks what happens in the periphery, while “periphery” knows the periphery, but also the centre, since “centre” is centre, and everyone has to relate to it. It will be a task for future research to study more closely when this periphery advantage mechanism strikes in Scandinavian political debate, resulting in a richer and more advanced grammar in public debates, and when interconnections between institutional and normative structures rather result in problematic biases and “lockins”.



Lessons for Normative Theory 

 131

7.4 Lessons for Normative Theory Summing up, the contributions of this chapter have been: to give an introduction to the standard academic debate on equal pay going on in philosophy, economics and certain branches of gender studies; to present the Norwegian equal pay controversy with its additional arguments both for and against equal pay and to give an interpretation of the particular profile and characteristics of the latter controversy and of how this controversy deviates from standard academic exchanges in light of insights from institutional theory. There is also a lesson to be drawn for academic normative theory and argumentation pertaining to how such scholarship relates to institutional theory and institutional variation. The obvious reasons for why academics and others deliberating on whether equal pay is a good idea or not need to acquaint themselves with scholarship on institutional variation, and empirical variation generally, are concerns of feasibility and implementability (Rothstein, 1994; Stemplowska, 2008; Elster, 2013): If the equal pay principle, or rather alternative principles, are to shape social practices, we need knowledge of such practices and the more detailed institutionalization of such practices to know how more concretely to go about in this or that society or polity. The discussions in this chapter suggest, however, an additional reason, namely that awareness of and inquiries into institutional variation and the nexuses between institutional and normative structures may be decisive for the strongest possible testing of normative arguments and principles. Different institutional templates spur typically different conceptualizations of relevant normative concerns and arguments, and as far as one is unaware of or uninterested in this variation important substantial argumentation could be overlooked and go unconsidered. The added arguments pro and contra equal pay from the Norwegian controversy may or may not turn out to be valid or tip the debate. They are, however, reasonable and seemingly relevant arguments, and a testing of the normative viability of the equal pay case that does not take them into account would seem to be weaker than a testing that does.

132 

 References

References Aaron, H. J. & Cameran M. L. (1986). The comparable worth controversy. Washington: The Brookings Institution. Acker, J. (1989). Doing comparable worth. gender, class and pay equity. Philadelphia: Temple University Press. Aldrich, M. & Buchele, R. (1986). The economics of comparable worth. Cambridge, Massachusetts: Ballinger. Barth, E., Moene, K. & Wallerstein, M. (2003). Likhet under press. Utfordringer for den skandinaviske fordelingsmodellen. Oslo: Gyldendal Akademisk. Barth, E., Røed M. & Torp, H. (2002). Towards a closing of the gender pay gap. Country Report Norway. Oslo: Institutt for samfunnsforskning. Blum, L. M. (1991). Between feminism and labor. The significance of the comparable worth movement. Los Angeles: University of California Press. Elster, J. (2013). Excessive ambitions (II). Capitalism and Society, 1, 3-34. England, P. (1992). Comparable worth. Theories and evidence. New York: Aldine de Gruyter. Evans, S. M. & Nelson, B. J. (1989). Wage justice. Comparable worth and the paradox of technocratic reform. Chicago: University of Chicago Press. Fraser, N. (2003). Social justice in the age of identity politics. In N. Fraser & A. Honneth (Eds.) Redistribution or recognition? A political-philosophical exchange. New York: Verso. Gunderson, M. (1994). Comparable worth and gender discrimination: An international perspective. Geneva: International Labor Office. Hammer, E. R. (1986). Pay inequity: A guide to research on social influences. New York: Garland. Hartmann, H. (Ed.) (1985). Comparable worth: New directions for research. Washington D.C.: National Research Council. Heath, J. (2009). Economics without illusions. Debunking the myths of modern capitalism. Toronto: Broadway Books. Hill, A. M. & Killingsworth, M. R. (Eds.) (1989). Comparable worth: analyses and evidence. Ithaca, N. Y.: ILR. Honneth, A. (2003). Redistribution as Recognition and The Point of Recognition. In N. Fraser & A. Honneth (Eds.), Redistribution or recognition? A political-philosophical exchange. New York: Verso. Høgsnes, G. (2002). Likelønn: Forhandlingsstruktur og makt. In A. L. Ellingsæter & J. Solheim (Eds.). Den usynlige hånd? Kjønnsmakt og moderne arbeidsliv, Oslo: Gyldendal Akademisk. Jacobsen, J. P. (1998). Policy application: Comparable worth. In J. P. Jacobsen (Ed.), The Economics of Gender. Malden, Massachusetts: Blackwell. Johansen, E. (1984). Comparable worth: The myth and the movement. Boulder, Colorado: Westview. Karamessini, M. & Ioakimoglou, E. (2007). Wage determination and the gender pay gap: A feminist political economy analysis and decomposition. Feminist Economics, 13(1), 31-66. Killingworth, M. R. (1990). The economics of comparable worth. Kalamazoo, Michigan: W. E. Upjohn. Krebs, A. (2003). Arbeit und liebe. Frankfurt am Main: Suhrkamp. Lapidus, J. & Figart, D. M. (1998). Remedying unfair acts: U.S. pay equity by race and gender. Feminist Economics, 4(3), 7-28. Macdonald, C. L. & Merrill, D. A. (2002). It shouldn’t have to be a trade: Recognition and redistribution in care work advocacy. Hypatia, 17(2), 67-83. Mahoney, J. & Thelen, K. (2009). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change. Cambridge: Cambridge University Press. Martin, C. J. & Swank D. (2011). Gonna party like it’s 1899. Party systems and the origins of varieties of coordination. World Politics, 63(1), 74-114.

References 

 133

Martin, C. J. & Swank, D. (2012). The political construction of business interests. Coordination, growth, and equality. Cambridge: Cambridge University Press. Martin, C. J. & Thelen, K. (2007). The state and coordinated capitalism: Contributions of the public sector to social solidarity in postindustrial societies. World Politics, 60, 1-36. Michael, R. T., Hartmann, H. & O’Farrell B. (Eds.) (1989). Pay equity: Empirical inquiries. Washington D.C.: National Academy. Milkman, M. (1990). Book Review: Ellen Frankel Paul, Equity and gender: The comparable worth debate. Public Choice, 66, 99-100. Mutari, E., Figart, D. M. & Power M. (2001). Implicit wage theories in equal pay debates in the United States. Feminist Economics, 7, 23-52. NOU 1997: 10 Arbeidsvurdering som virkemiddel for likelønn NOU 2008: 6 Kjønn og lønn. Fakta, analyser og virkemidler for likelønn Hearing reports to NOU 2008: 6. See http://www.regjeringen.no/nb/dep/bld/dok/hoyringar/ hoeringsdok/2008/horing---nou-2008--6-kjonn-og-lonn/horingsuttalelser.html?id=505160 O’Neill, J. E. (2000). Comparable worth. See http://www.econlib.org/library/Enc1/ComparableWorth.html Paul, E. F. (1993). Equity and gender. The comparable worth debate. London: Transaction Publishers. Pincus, L. & Shaw, B. (1998). Comparable worth: An economic and ethical analysis. Journal of Business Ethics, 17, 455-470. Rhoads, S. E. (1993). Incomparable worth: Pay equity meets the market. Cambridge, United Kingdom: Cambridge University. Roback, J. (1986). A matter of choice: A critique of comparable worth by a sceptical feminist. New York: 20th Century Fund. Rothstein, B. (1994). Just institutions matter: The moral and political logic of the universal welfare state. Cambridge: Cambridge University Press. Schwab, D. J. (1985). Job evaluation research and research need. In Heidi Hartmann (Ed.), Comparable worth: New directions for research. Washington D.C.: National Research Council. Skjeie, H. & Teigen, M. (2003). Menn imellom. Mannsdominans og likestillingspolitikk. Oslo: Gyldendal Akademisk. Sorensen, E. (1994). Comparable worth? Is it a worthy policy? Princeton, New Jersey: Princeton University Press. Stemplowska, Z. (2008). What’s ideal about ideal theory? Social Theory and Practice, 34(3). Tenden, E. B. (2001). Hvorledes likestilling? En analyse av konstruksjonen av likestillingslovens bestemmelser om likelønn, arbeidsplikt og seksuell trakassering 1990-2001. Hovedoppgave, Institutt for statsvitenskap, Universitetet i Oslo. Thelen, K. (2009). Institutional change in advanced political economies. British Journal of Industrial Relations, 47, 471-498. Treiman, D. & Hartmann H. (Eds.) (1981). Women, work and wages: Equal pay for jobs of equal value. Washington D.C.: National Academy Press. van der Vleuten, A. (2007). The price of gender equality. London: Ashgate. Weinzweig, M. (1987). Pregnancy leave, comparable worth, and concepts of equality. Hypatia, 2(1), 71-101. Wilborn, S. L. (1986). A comparable worth primer. Lexington, Massachusetts: Lexington Books.

PART II: Institutional Change in Work and Vocational Education

Dag Olberg

8 Regulating the Temporary Layoff Institution – Coalitions and Drift Tripartite cooperation and concertation is often used to characterize the Norwegian labour market model. The changing regulations concerning temporary layoffs illustrate one characteristic which is less highlighted in the literature: bipartite coalitions within the tripartite cooperative model. In an institutional perspective, the discussions and struggles concerning the regulation of temporary layoffs illustrate both a special case of institutional drift and that bipartite coalitions within the tripartite model may provide a source of stability. The long lasting temporary layoff institution illustrates one of the special traits of Norwegian labour market regulations. Temporary layoffs may be used during economic downturns and in some other situations where a company has no work for its workers. The affected employee is not expected to fulfil the obligations according to the employment contract. The employee receives unemployment benefits supported by the state, not ordinary wage, yet is at the same time not regarded as laid off, as the employment relationship still exists. The temporary layoff institution is not a general support scheme; it is designed to support otherwise economically viable companies in shorter periods of unforeseen lacking demand and critical trade cycles. As background, the chapter addresses the changing regulations underpinning this institutional setup in different labour market situations; one is the economic downturn following the international financial crisis from late 2008, and the other concerns the recovered and stable Norwegian labour market in 2012. Research questions concern the role of the social partners in particular, as the temporary layoff institution is historically based on collective agreements, yet also regulated by law. The chapter addresses the use of temporary layoffs in a European and Nordic context, and seeks, in relation to the Norwegian social partners, to explore the construction of interests and the category of drift in a coalitional perspective.

8.1 Introduction Although the effects of the crisis in Norway were comparatively mild, parts of the open Norwegian economy were immediately affected. Initially, this mainly concerned producers in export industries, but the effects soon spread to supplier industries and building and construction companies. Other sectors were also affected. The economic crisis-packages commissioned by the state, first in order to restore and secure the functioning of the financial system and later in order to promote activity in the building and construction industry, were the most important measures taken. Yet, in common with experiences in other countries, © 2015 Dag Olberg This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

Introduction 

 137

in several companies the effects of the crisis had already led to over-capacity and job-losses, which brought working time issues on the agenda. In this context the chapter addresses the extended use of temporary layoffs during the initial phase of the economic downturn in 2008/2009 and after the, comparatively quick, recovery in 2011/2012. In both the literature (Evju, 2007; Hansen & Kvadsheim, 2010; Olberg 2010, 2013) and in official labour market documents (NOU, 2013, p. 13), temporary layoff arrangements are referred explicitly to as an “institution” – with a history and tradition and with regulations and practices. It was formalized first in the late 1940s in the Basic Agreement between the LO (confederation of labour unions) and the NHO (the main employers’ confederation), yet different forms of temporary layoffs were also practised earlier, in the 1930s and 1920s, for instance in the fish processing industry in periods with lack of incoming raw material (Evju, 2007; Opsal, 2007). At the present time temporary layoffs are regulated at different levels: collective agreements, law, and customary law.

8.1.1 Chapter Outline The chapter is divided into four sections. First is a brief presentation of research questions followed by a short overview of data. Section two consists of background presentation relating to working time and employment issues in the context of the economic downturn resulting from the financial crisis. This short overview of European comparative and institutional perspectives is followed by a brief Nordic comparison. Next is a presentation of the main characteristics concerning the Norwegian temporary layoff institution and an overview of the extent of temporary layoffs in the period from August 2008 till December 2012. Section three presents regulatory changes relating to the temporary layoff institution, with a special focus on the role of the social partners. This part gives an account of regulations and re-regulation. Although this is a continuing story, the proceedings of autumn 2012 may pose some questions regarding institutional change, regulation and the role of strategic actors. Section four discusses the temporary layoff case in light of institutional perspectives; this section also discusses the construction of interests from a political coalitional perspective. In the conclusion I discuss the regulation and re-regulation of the temporary layoff institution as a special case of institutional drift.

8.1.2 Research Questions The main empirical research questions posed here concern the role of the social partners in their involvement with temporary layoff regulations during the economic downturn in 2008/2009 and in the relatively stable (more accurately, growing)

138 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

Norwegian labour market of 2011/2012. Which steps of action were taken by the actors, which were the reasons and how are we to interpret these actions in the context of institutional change? And, to what extent may the case presented shed light on the construction of interests? Several perspectives and approaches exist within institutional theory, also regarding the possibilities of integrating different institutional perspectives– in particular historical-, sociological-, and rational choice institutionalism (Hall, 2010; Mahoney & Thelen, 2010; Thelen, 2012). Relating to the particular type of institution presented in this paper, I discuss the Norwegian temporary layoff institution in a coalitional perspective. The discussion also addresses Nordic variations regarding the use of temporary layoffs and shorttime working during the crisis, which also illustrates similarities and differences between the Nordic models. The discussion of the temporary layoff institutions also provides an opportunity to address questions concerning interest construction (Martin & Swank, 2012) and institutional drift (Streeck & Thelen, 2005; Mahoney & Thelen, 2010). Beyond mere description and reporting of the case story (regulatory changes and the use of temporary layoffs), is there anything which should be accounted for in terms of explanations? I would argue that there is, as the controversies over regulations of autumn and winter 2012 were not to be expected. The story of how this strife was handled and eventually brought to an end (so far) also sheds light on the stability of this type of institution and the way it changes during and after economic crises. More generally, this chapter discusses which role coalitions of actors may have in influencing institutional regulations and practices. I also argue that it is of general interest to explore to what extent a coalitional perspective may expand the understanding of the category of institutional drift.

8.1.3 Data The data consists of laws and regulations concerning temporary layoffs, regulations in collective agreements and written communications by the social partners during the period (see also Olberg, 2010, 2013). I also draw on media coverage to illustrate how the public discussion proceeded. Statistical data on unemployment and the scale of temporary layoffs have been gathered from the labour market statistics presented monthly by The Norwegian Labour and Welfare Administration (NAV).



The Crisis – Overcapacity, Working Time and Employment Relations 

 139

8.2 The Crisis – Overcapacity, Working Time and Employment Relations 8.2.1 European Comparative Institutional Perspectives Relatively large differences exist between European countries regarding weekly working hours, working time in different sectors, statutory maximum working time, holidays, annual working time, etc. Yet, a common trait is that few steps have been taken to reduce weekly working time. Data covering the EU-15 countries and Norway show that during the last 10 years the average weekly working time was reduced only from 38,6 hours to 37,9 hours – a reduction close to two per cent during the period. Most of this reduction took place before 2003, and in most countries working time reductions without reduction in pay have hardly been a collective bargaining issue. Working time reductions with reductions in pay, however, have recently been on the agenda in many countries as the effects of the financial crisis resulted in falling demand and accompanying over-capacity in many companies (Eurofound Eiro, 2009; see also Glassner & Galgóczi, 2009). It has been noted that the social partners have played an important role in implementing statutory short-time working provisions, yet there are also questions to be raised regarding the companies’ need for working time flexibility in restructuring and the role of collective agreements. One concerns what factors enhance or hinder negotiated responses by the social partners, another concerns the role of state policies and measures, linked to collective bargaining and a third issue concerns the character of the negotiated responses in terms of contents and measures included in collective agreements (Glassner & Keune, 2010, p. 6). The heterogeneity of the European labour markets and the diverse state policies has also been addressed in relation to working time and the economic downturn following the financial crisis. Andersen (2009) noted that short-time work so far appears to be one of the most prominent tools to prevent dismissals and that the various forms of regulation of short-time work to some extent mirrors the different industrial relations regimes in Europe. While a liberal approach with relatively few state-financed subsidies characterizes the UK and some of the Eastern European countries, states more often intervene in the labour market in the Western part of continental Europe. In Scandinavia there are mixed patterns of regulations, as employers’ associations and trade unions have relatively large autonomy in collective bargaining, while labour market policies and subsidy schemes financed by the state also play a central role (Andersen, 2009, p. 5). It has been noted that in most countries companies do not have access to temporary layoffs within the institutional framework (Hansen & Kvadsheim, 2010), yet it is clear that in a number of countries (Austria, Belgium, France, Germany, Italy, the Netherlands) governments have extended the use of already existing labour market schemes for handling seasonal fluctuations (Glassner & Galóczy, 2009). Glassner

140 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

and Keune (2009) report that in countries such as Bulgaria, Hungary, Poland and Slovenia, legal provisions for short-time work have been introduced only recently. As for the Nordic countries, although illustrating important similarities regarding labour market models and tripartite cooperation, several differences exist regarding access to, and the form of, short time work and temporary layoffs. As discussed in Section 4 below, these differences were spelled out during the period 2008/09 – 2012.

8.2.2 Nordic Comparison The Nordic labour market models expose several common characteristics. However, prior to, and during, the financial crisis there were substantial variations between the Nordic countries regarding institutional set ups facilitating short-time work and temporary layoffs (Andersen, 2009; Svalund et al., 2013). Both Norway and Finland had established schemes for temporary layoffs, while work-sharing schemes had existed in Denmark for several years. Although these schemes also varied, common characteristics were that they involved state subsidies (unemployment benefits) and that costs also were shared between companies and workers. Sweden was the deviant case, as there was no tradition for work-sharing and no temporary layoff institution comparable to the ones in Norway and Finland. Yet, local agreements on job security and training financed by the employer existed in a number of Swedish companies. In Sweden there was no legislation concerning wage subsidies regarding shorttime work or temporary layoffs, yet in 2009 an agreement was reached between the metal workers union and the employers on a (maximum) 20 per cent cut in pay if working time was reduced by one day or more (Andersen, 2009, p. 6). Concerning Denmark, Andersen notes that social security contributions linked to short-time work have been part of Danish labour market regulation for many years, yet the government, after having extended the period of coverage from 13 weeks to 26 weeks in spring 2009, rejected claims of extended duration – although pressure from the major industrial employers existed (Andersen, 2009, p. 7). The Norwegian temporary layoff institution seemed somewhat more generous and in the context of the recent crisis, perhaps, more flexible. As response to the economic downturn in 2008/2009 regulations concerning state funding of temporary layoffs were swiftly relaxed in Norway and Finland. Among these regulatory changes, the extension of the period employees on temporary layoff were entitled to receive unemployment benefits was most important. The Danish response was more mixed as the maximum work-sharing period was in 2010 reduced from four to two years and the criteria for re-qualifying tightened (Jørgensen & Schulze, 2011), yet the distribution of work periods and unemployment periods was relaxed in 2009 (Jørgensen, 2011). Danish employers required an extension of the work-sharing



The Crisis – Overcapacity, Working Time and Employment Relations 

 141

period from 26 weeks to 18 months, a claim which was supported by the unions but not conceded to by the centre-right government. As there was neither a temporary layoff institution nor work-sharing schemes in Sweden, it was seen as something new when the metal workers’ union and employers in manufacturing signed an agreement on how to handle job losses at company level. This included working time reductions (to 80 per cent) without compensation and/or days off with corresponding pay reductions (Kullander & Henrikson, 2010). In addition to temporary layoffs, the central agreement also opened up for local agreements on training and upgrading of skills.1 The peak union and employer confederations seemed to be most influential in Norway, while in Finland the debate was more limited. In Denmark the government did not concede to the unison claims from the union and employer side regarding extending the work-sharing period, so the employment coalition was not as strong as in Norway and did not succeed in this respect. In Sweden the government did not support temporary layoffs financially, and the parties on the employer and union side had to cope themselves (the 2009 collective agreement contributed to saving jobs but at the cost of wage reductions). Importantly, in Norway (and Finland, but not in Sweden and Denmark) the temporary layoff institution already existed. The social partners and the state had had experiences from using it during previous economic downturns and crises, and it could readily be put to use. It may also have played a role that the political settings were different, with a social democratic-led government coalition in Norway, conservative governments in Sweden and Denmark and a centre-conservative government coalition in Finland. Although the Nordic industrial relations systems share several traits (among others various forms of coordinated two-tier bargaining), there are also important variations. Industrial relations in Norway are more centralized compared to the other Nordic countries, although union density is lower. The social partners in the Nordic countries established early on basic agreements regulating mutual recognition, coordinated collective bargaining, mediation and labour peace during periods between the bargaining rounds. The Norwegian Basic Agreement was signed in 1935, after a period of harsh labour conflicts, including wide lock-out actions from the employer side. The Swedish Saltsjöbaden Agreement was signed in 1937 (Lundh, 2010), and the Danish September accord as early as 1899 (Due & Steen Madsen, 2010). What characterizes the Norwegian model compared to the

1 Stål och Metall Arbetsgivareförbundet, Gruvornas Arbetsgivare SVEMEK / Industrifacket Metall (2009) Protokoll: Tillfällig permitteringsöverenskommelse, § 5 Utbildning. The agreement was prolonged and revised in 2012, after which government in 2013 forwarded a legislative proposal including a public scheme for short-time work to be used only in instances of deep recession and not preventing necessary structural labor market changes (Svalund et al., 2013).

142 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

other two is the relatively strong and lasting degree of centralization and coordination, a stable climate of cooperation (different from Sweden) and tripartite coordination including a central and more active role played by the state (different from Denmark). It has been noted that while the “spirit” of the Swedish Saltsjöbaden Agreement was weakened by the deterioration of industrial relations from the 1970s, the “spirit” of the Norwegian Basic Agreement still exists (Bergh, 2010, pp. 14-15).

8.2.3 The Norwegian Temporary Layoff Institution In Norway the tradition of temporary layoffs goes back to the 1920s, possibly even earlier in the fishing industries (Evju, 2007, p. 147). In 1947 it was agreed by the employee- and employer confederations, LO and NAF (now NHO), to regulate the temporary layoff practice in the context of the Basic Agreement. The intent of the arrangement was to avoid layoffs in companies that in short periods of time experienced lack of demand. In this context the agreement regarding temporary layoffs was initially a case of bi-partite cooperation. A special characteristic of temporary layoffs is that the employee is fully or partly unbound by the obligation to perform work according to the employment contract. Likewise, the employer is freed from the obligation to pay wage. Yet, the employment contract is maintained; the employee is not dismissed. The arrangement has been described as a “hybrid” form of employment (Hansen & Kvadsheim, 2008). Initially a bipartite arrangement, the state did, after some time and according to shifting rules and regulations, take over most of the economic responsibilities by supporting the temporarily laid off workers via unemployment benefits. Thus the temporary layoff institution illustrates one important aspect characterizing the Norwegian model, as the burdens are shared by the state, the employer, and the employee. It also illustrates one aspect of the institutionalized compromises between class interests, yet the temporary layoff institution has also been debated (among others, see Torp, 1987; Nicolaisen & Lismoen, 2004; Nergaard, 2010). Issues in these discussions have been financial costs and economic responsibilities (the use of social security contributions, employers wage obligations), labour market politics (counter-measures in economic downturns, lock-in-effects) and industrial policies (in Norway also regional policies and local politics). The institution of temporary layoffs itself is not, and has not been, based on law, yet it is imbued by regulations and after some time also regulated in law provisions concerning the use of state unemployment benefits for temporarily laid off employees. The LO/NHO Basic Agreement contains a quite broad set of regulations on procedures, which are also referred to in collective agreements, among others the right to early warning and the right for shop stewards to consultations.



Regulatory Changes – Variations According to Labour Market Situation  

 143

Local employer

Employee

State, The Norwegian Labour and Welfare Administration (NAV)

Period with wage obligations for employer

”Waiting days” Reduced (no) income

Unemployment benefits, according to rules and regulations

Figure 8.1: The Norwegian temporary layoff institution, distribution of costs.

Public regulations include law regulations and other regulations on how social security contributions may be spent in situations with temporary layoffs, including obligations for employers to pay wage for a period of time after temporary layoff has been decided, and they set the number of “waiting days” where the employee receives neither pay nor unemployment benefits. Public regulations also set rules stating that in situations where temporary layoffs are partly used, the reduced working time should amount to a certain percentage (50 per cent before the last changes in regulations spring 2009, then 40 per cent). Unemployment benefits normally cover just above 60 per cent of ordinary pay.

8.3 Regulatory Changes – Variations According to Labour Market Situation The use of temporary layoffs varies according to business cycles and labour market situation; in addition there are seasonal variations in some industries (such as the fishing industries and parts of the building and construction industries).

8.3.1 The Ebb and Flow of Regulatory Changes In the economic downturn of 2003/2004 temporary layoffs were relatively common. The start of a new upturn in 2004 was followed by new debates on the temporary layoff regulations, as the period of wage duty for the employer was extended and the period of time a temporary laid off person was entitled to receive unemployment benefits was shortened (Nicolaisen & Lismoen, 2004). Regulations of temporary layoffs have been changed several times, changes relating to both downturns and upturns in business cycles and to labour market problems affecting different industries. Some of these changes – before, during and after the crisis – are illustrated below.

144 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

2004

2009

2012 - 2013

Employers’ wage duty period

Increased from 3 to 10 days

Reduced to 5 days

Increased to 20 days

Employees’ “waiting days”

5 days

3 days

3 days

52 weeks

26 weeks

Period with 30 weeks (was 52 unemployment benefits weeks until 2002)

Figure 8.2: Regulatory changes in temporary layoffs.

The re-regulation of February 2009 was one of several countermeasures against the economic downturn. Regulations concerning employers’ wage-duty period, employees’ “waiting days” and the period when a temporarily laid off employee was entitled to receive unemployment benefits were swiftly relaxed. Also, among other initiatives, in March 2009 initiatives were taken in order to ensure that also temporary layoffs, where the employee remained on 60 per cent of full time, could be entitled to unemployment benefits – a shift that facilitated rotating layoffs among the employees.

8.3.2 Regulations 2009 – The Role of the Social Partners As noted above, in early 2009 the government decided to relax parts of the regulations covering the use of state unemployment benefits when companies decided temporary layoffs were necessary. The justification was the need for the temporarily laid off employees to secure income and companies’ need to retain the competence and skills of their workforce. The changes in regulation were initially suggested by the social partners. Having received several inquiries from both local union representatives and companies, the central heads of the employee- and employer confederations (LO and NHO) joined forces, writing a letter to the Prime Minister requiring more flexible regulations regarding the state’s role in financing employees on temporary layoffs. The government complied with several of these requirements. This implied that some of the regulations covering temporary layoffs in the Basic Agreement also had to be changed. This was agreed upon by the central social partners in a protocol of early June 2009. The initial requirements regarding employers’ wage duties were somewhat more extensive than what the government had decided.2 The employers’ industry federation, Norsk Industri, noted that in principle it was possible to apply temporary layoffs with even shorter working time, for instance 20 per cent, but this was unrealistic as the employees in such cases would not be entitled to unemployment

2 Requirements were four days wage duty with 80 per cent layoffs, three days wage duty with 60 per cent layoffs, two days wage duty with 40 per cent layoffs (nho.no, 4.2.2009).



Regulatory Changes – Variations According to Labour Market Situation  

 145

benefits, and the employer’s wage duty period was seen to be too long (norskindustri. no, 25.3.2009). The employer federation in the building industries welcomed the government’s changes of regulations (bnl.no, 24.3.2009). Several of the regional offices of the NHO employer confederation had earlier pressed for more flexible regulation of temporary layoffs. Yet the NHO at the central level had been reluctant to some of these claims. The NHO, understanding that local employers wanted regulations matching the needs of their own companies, in January underscored that regulations should match the needs of business in general. Rather than extending the maximum time period for using temporary layoff (at that time about half a year), more flexible regulations were required (shortening the period with wage duty, making it easier to rotate employees on temporary layoffs, etc.) It was also noted that extending the total length of temporary layoffs would restrict employers in other companies and industries from making use of the work force of employees who had been temporarily laid off (DN 21.1.2009). Eventually, as discussed above, the outcome during spring 2009 was both extended time in which temporary layoffs supported by unemployment benefits could be applied and somewhat more flexible regulations. Comments from the LO-union for industrial workers, Fellesforbundet, were in general somewhat less sceptical. Regarding temporary layoffs, the union noted that it was advantageous that employees still were covered by an employment contract and that temporary layoffs were a better alternative than ordinary layoffs (fellesforbundet. no 3.3.2009).

8.3.3 The Extent of Temporary Layoffs By the end of February 2009 the number of unemployed persons was 67  400, an increase of 25 400 persons compared to the previous year. Unemployment was highest among men and young people and in the building and construction industries. In the economic downturn of 2008/2009 common responses of companies facing overcapacity were layoffs, temporary layoffs or other means of reducing staff (not replacing staff in cases of retirements, transition to voluntary pension schemes, reducing the use of hired personnel, temporary agency work, etc). Reports from The Norwegian Labour and Welfare Administration (NAV) showed that many firms used a form of temporary layoffs where the employee worked reduced hours. In February 2009 that applied to four out of 10 among the 12 500 temporary laid-off employees. The suggestion is that the possibility, according to rules and regulations, to work reduced hours while being on temporary layoff schemes, increases firms’ flexibility in economic crises. In the period from August 2008 till august 2009 the number of temporarily laid off persons reached a peak in April 2009, with a total of more than 18 000 employees on temporary layoff. Registered unemployment in September 2009 was down to 2.7 per cent (NAVfigures, 69  600 persons) of the total work force. That the number of persons on

146 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

temporary layoffs also decreased, which suggested that the economic down-turn (in Norway) had milder effects than expected but also that the counter-measures had effects relatively quickly. These counter-measures were not only targeting the banks and the financial sector, they also included funding public demand (building projects), a range of labour market schemes and, as noted above, changes in regulations covering the temporary layoff institution. Yet, the number of employees on temporary layoff again increased during winter 2010. The use of temporary layoffs was rapidly extended from autumn 2008. In February 2009 about 18 000 persons were temporarily laid off, fully or partly; a year later this number was close to 20 000. In the second half of 2012 less than 5000 persons were registered as temporarily laid off – nearly down to the level before the financial crisis. Total

Temporary layoffs

Temporary layoffs on reduced working time

25000 20000 15000 10000 5000

Oct

Dec

Jun

Aug

Apr

Feb

Oct

Dec

Jun

Aug

Apr

Feb

Oct

Des

Jun

Aug

Apr

Feb

Oct

Dec

Jun

Aug

Apr

Feb

Oct

Dec

Aug

0

Figure 8.3: Temporary layoffs in the period August 2008 – December 2012 (Source: based on figures from The Norwegian Labour and Welfare Administration NAV).

8.3.4 Regulations and Re-Regulation – 2012 The relaxed regulations of 2009 were in effect during both 2010 and 2011. Both the LO and several industry-based unions and the NHO employer confederation with its affiliated industry organization Norsk Industri had warned against tightening regulations – as would be the usual practice when production and labour market were back to normal. Both the employee and employer sides argued that due to the financial crisis and the debt crisis the economic situation in Europe was still uncertain, something that would also affect exposed sectors in Norwegian industry. These arguments were accepted by the government, and the temporary layoff regulations were prolonged. Only in 2012 were the regulations tightened. The Minister of Labour noted that labour market conditions had improved and that the relaxed regulations of 2009



Regulatory Changes – Variations According to Labour Market Situation  

 147

no longer were needed. As a consequence the changes introduced in connection with the financial crisis of 2008/2009 were reversed. This made it more expensive for companies to make use of the temporary layoff institution. Preparing the 2013 state budget (Meld. 1. St. 2012-2013), the government noted that the economic growth seemed to continue, with growing productivity, increased labour market participation and reduced unemployment. The restrictions introduced in 2012 were prolonged. Both the union Fellesforbundet (and other unions) and the employer organization Norsk Industri had during autumn 2011 opposed the proposed changes in the 2012 budget. Reactions concerning the role of the temporary layoff institution in 2013 were much less pronounced. All in all, the regulatory restrictions seemed to have been accepted.

8.3.5 Change in Practices – Unemployment Benefits and Temporary Layoffs Drift, as one type of institutional change, is defined as the changed impact of existing rules due to shifts in the environment (Mahony & Thelen, 2010, p. 16.) It tends to characterize situations with a high level of discretion in interpretation/ enforcement on the one hand and with strong veto possibilities for actors on the other. Drift situations may be characterized by gaps between rules and enforcement (Mahony & Thelen, 2010, pp. 19-22). Drift seems to be a useful category in order to understand the struggles surrounding the temporary layoff institution. Controversy arose over the way the Norwegian Labour and Welfare Administration (NAV) administered unemployment benefits in temporary layoff situations. Several industry unions received complaints from local union representatives that temporarily laid-off workers did not receive unemployment benefits as they would have according to earlier practice. Such situations were both temporary layoffs due to seasonal variations (winter/cold weather) and market fluctuations (companies temporarily in difficulties due to lost contracts, etc.). The more restrictive regulations implied that unemployment benefits in connection with temporary layoffs would only be agreed upon if the situation could not have been foreseen by the employer and the situation was one the employer could not influence. Seasonal variations, NAV argued, were foreseeable;3 likewise, market

3 One Secretary of State, representing the social democratic political leadership in the Ministry of Labour, noted in autumn 2011 that the regulations concerning the use of layoffs due to seasonal variations would not be relaxed. Employers engaged in road construction could not base their business on temporarily laying off their employees a couple of months every winter, expecting the state to support their workers with unemployment benefits during the period. (Arbeidsmanden 10/2011, pp. 14-15). Seasonal variations are foreseeable for employers, and in Norway cold weather is foreseeable. Yet, it was also noted that the practices of the Norwegian Labour and Welfare Authorities differed in the various regions before the tightening of regulation practices, and it was hoped that the common national practices would now be established.

148 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

fluctuations and lost contracts should be counted as part of ordinary commercial risks. The more restrictive practices were strongly opposed by the unions, first at the local level. Several incidents were reported – for example, regarding companies who lost building contracts to foreign companies that based their activities on the use of cheap foreign labour (see also Nicolaisen & Trygstad’s chapter). The discontent resulted in a planned political protest strike against NAV’s restrictive practices in the southwestern part of the country. At the central level both the LO and NHO reacted against the more restrictive practices, the LO most strongly. The LO warned that the new practices could lead to ordinary layoffs and in addition more widespread use of temporary employment contracts. The NHO noted that earlier, if temporary layoffs were agreed upon locally between the company and the union representatives in accordance with the Basic Agreement, this would normally be sufficient for NAV to grant unemployment benefits to the affected workers. As a consequence representatives from the LO and NHO at the central level met with the Minister of Labour, who made the Labour and Welfare Directorate reconsider their practices, while the LO and NHO were invited to participate in further work concerning the issues. This was in practice tripartite cooperation. In mid-November 2012 the directorate issued temporary guidelines instructing the local NAV offices to implement less restrictive practices. If the social partners agreed locally that the employer had made the provisions that could reasonably be provided and that the reason for the temporary layoffs was unforeseeable, this should normally guide NAV’s decisions.

8.3.6 The Accord The new permanent guidelines from the directorate (in force from February 2013) illustrated that the LO’s and NHO’s protests had won through. When a temporary reduction of production occurs, the use of temporary layoffs is regarded as one of the measures an employer may use as a flexible adjustment to the market situation. The new guidelines list as possible reasons for applying temporary layoffs according to the Basic Agreement and customary law illustrations such as lacking orders, full stores, lack of raw material, the financial crisis, loss of tenders, and so on (Del A, pkt. 4). Regarding the payout of unemployment benefits, the NAV offices should, as a general rule, base their decisions on the local parties’ agreement on whether the situation suggested temporary layoffs or ordinary layoffs. The restrictions regarding seasonal variations were largely not changed. Seasonal variations were seen as something the employer could foresee and adjust to. Both the LO and the NHO approved the new guidelines. By the start of 2013 common agreement on the new guidelines was stated, and the announced political strike actions were called off.



The Temporary Layoff Institution and the Role of the Social Partners  

 149

In spite of the controversies and heated debate, the process and the changes made were – afterwards – described by all involved parties as an example of constructive tripartite cooperation.

8.4 The Temporary Layoff Institution and the Role of the Social Partners The evolvement of the Norwegian temporary layoff institution illustrated how questions relating to effective labour allocation and possible lock-in effects are being balanced against maintaining labour relations in ways that are affected both by tradition and the power of political coalitions and norms. Such considerations were also basic in other Nordic countries, yet the events following the financial crisis illustrated differences regarding both the use of temporary layoff schemes and the role of the central social partners.

8.4.1 Political Coalitions – Bipartite Coalitions Within the Tripartite Model In the Norwegian case, the actions initiated by the LO, and supported by the NHO in autumn and winter 2012 illustrated a bipartite coalition within the tripartite cooperative model. This is a characteristic less highlighted in the literature concerning the Norwegian model – where cooperation and trust between the three parties is often underscored. Yet, that two actors in the threesome form alliances and mobilize their resources in opposition to the third party is nothing new. In 2005 the employers were relatively satisfied with the centre-right government’s proposals to change the Work Environment Act, changes the trade unions opposed. However, the changes were reversed by the new coalition government led by the Labour Party after the elections in 2005, something which was welcomed by the unions and the LO. In 2006, however, the LO and NHO joined forces, as they strongly opposed the government’s planned changes regarding financing of the sick pay scheme (leaving it to enterprises to pay a somewhat greater share in the early phase of an employees’ sick leave; see also Hagelund & Pedersen’s chapter). One argument was that such changes, according to the Inclusive Working Life Agreement, should be subject to negotiations. A more recent illustration is from 2011, when the NHO opposed the government’s whitebook on work environment and regulatory life issues (St. Meld. 29 2010-2011). The NHO noted that it was unfortunate that this central document was presented in close consensus with the LO only, as this, according to the employer confederation, put the tripartite cooperation under considerable pressure. In 2012, as described above, it was the LO and NHO which joined forces in the dispute concerning the temporary layoff regulations. However, following the election in 2013 the new conservative

150 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

government suggested several regulatory changes concerning working time and temporary employment contracts. This time the changes were welcomed by the NHO but opposed by the LO. Tripartite cooperation and concertation is a characteristic trait of the Norwegian model, yet, so are shifting coalitions and repeated games. The Norwegian labour market model, it can be argued, is not mainly based on trust; rather, it is based on compromises – and such compromises may over time strengthen trust. Dølvik (2013) has argued that, by and large, the actors of the Nordic models have mostly succeeded in agreeing on compromises that have over time been sufficiently coherent in order to handle upcoming problems. This includes looking after crossing concerns and interests in ways that are sufficiently balanced in order to reshape actors’ legitimacy, rating, trust and willingness to participate in new rounds of pragmatic and coordinated problem-solving (Dølvik, 2013, p. 70).

8.4.2 Interest Construction – Mobilization and Power, Renegotiations In this case, the interests and motives of the parties are shaped by a long history of negotiations and re-regulation and also by the bargaining power, yet also by a relatively limited set of opportunities, resulting from bipartite and tripartite coalitions.4 A basic motivation among employers was related to skills and competences. Trade unions were motivated by solidarity and equality (Olberg, 2010; Svalund et al., 2013). As the employment contract was still binding, the employers were able to retain skills and competences among their employees. As argued in the VoC literature employers in coordinated market economies regard skills as a competitive advantage and seek to retain skilled employees.5 The unions, on their side, were interested in job security, if it could be attained, but also in regulations that made it possible to divide the burdens more equally between the employees. An additional reason why the peak organizations on the trade union side defended (their conception of) the temporary layoff institution is related to the fact that it concerns the unions’ core constituencies. In the economic downturn following 2008/2009 the first groups affected were employees on short-term contracts and agency-workers. The temporary layoff institution is, like the German Kurzarbeit schemes, not re-distributive; such schemes are mostly for the core workers and the core labour market – that is, in the Norwegian context, primarily for organized labour. Insiders benefit, and that is one important reason why the LO wants to control this institution, as it privileges some constituencies and not others. Yet,

4 For a broader analysis of the political construction of business interests, see Martin & Swank, 2012. 5 “Since many firms in coordinated market economies make extensive use of long-term labor contracts, they cannot rely as heavily on the movement of scientific or engineering personnel across companies, to effect technology transfer, as liberal market economies do” (Hall & Soskice, 2001, p. 26).



The Temporary Layoff Institution and the Role of the Social Partners  

 151

temporary layoffs may also be used by companies without collective agreements or trade union representatives. However, in such companies the procedural regulations on temporary layoffs in the Basic Agreement are not necessarily applied (resulting grievances are part of the background why some labour lawyers and unions outside the LO recommend that the temporary layoff institution should be regulated by law, also providing unorganized workers with procedural rights which has traditionally been reserved for organized labour).6 This aspect of interest formation on the union side was also two-sided. The interests of the LO were also shaped by a strong wish to avoid what was seen as a possibly extended dualism of a more segmented labour market based on employers’ possible extended use of dismissals in combination with temporary work contracts. The NHO also wanted to maintain the temporary layoff institution but simultaneously lamented the lack of other flexibility measures, such as the use of temporary work contracts and other types of flexible use of labour. It was also important to the bipartite coalition that the arrangement was financed by the third part, the state, which paid out the unemployment benefits. The state authorities, however, were also concerned with possible lock-in effects and labour market flexibility in general. In this perspective they were also concerned with institutional drift, in the sense that effective labour allocation would be hindered if the temporary layoff institution developed into a more general support scheme. The temporary layoff institution changes and adjusts to shifting labour market conditions – this is part of the institution’s “rationale”. The central social partners accept these adjustments (even though they may oppose, and warn against, restrictions before they are introduced). Yet, as illustrated by the events during late 2012, there are limits. The trade union and employer confederations are rational actors; they mobilized their resources and actively pursued their interests. In interpreting the actions taken regarding temporary layoffs, a rational choice perspective seems basic. The temporary layoff institution is not a self-contained social structure but is actively supervised and looked after by the social partners. This seems obvious, but there are also other aspects to the regulation of the temporary layoff institution. The controversies and heated debates during autumn 2012 were, basically, not on the regulatory changes themselves but on practices. The new practices first spread from some regional NAV offices. The dispute did not concern the length of the temporary layoff period, which had been reduced,

6 Supporting this argument, it may be added that the LO has resented claims that employees’ procedural rights to codetermination in temporary layoff processes should be granted by law provisions, arguing that the procedures stated in the Basic Agreement suffice, and that if employees in companies without collective agreements should want to be covered by such protection they should be organized and covered by collective agreements rather than seek protection via law provisions.

152 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

but the use of discretion in deciding whether the affected workers were entitled to unemployment benefits at all. Again, following my interpretation, this was regarded as something new by both trade unions (in particular) and employers – something which, according to a logic of appropriateness (March & Olsen, 1984; Scott, 1995, p. 39) was seen as unfitting, unwanted and even provocative. In 2012 the employer and trade union confederations did not object to the view that effective allocation of labour was in general an important issue. They had, however, a different definition of the labour market situation and also different opinions and definitions regarding the temporary layoff institution itself. One illustration is that local shop stewards were advised by their union not to accept and not to collaborate in local temporary layoff schemes if it was not clear that their members would receive unemployment benefits. Another issue which was discussed on the trade union side was that the new situation in autumn 2012 could make it necessary for the LO and NHO at the central level to renegotiate and rewrite the clauses concerning temporary layoffs in the Basic Agreement. As noted above, there had also been announced a political strike in order to make NAV revise practices. All in all, these may be interpreted as quite strong measures. One obvious reason for the trade unions’ negative reactions is that workers (and trade union members) that would earlier have received unemployment benefits were in some cases left without income. Additionally, and importantly, the trade unions feared that NAV’s new practices would lead to an extended use of non-standard work contracts and employment relations, such as temporary contracts and temporary agency work. For the employer side, temporary layoffs had traditionally been a flexible buffer. Temporary layoffs were preferable to ordinary layoffs, as the use of such schemes made it possible to retain important competency in the workforce also in periods with unforeseeable fall in demand. Additionally, it mattered that the trade union- and employer confederation regarded the temporary layoff institution as, in a special way; “their” institution. It had originated many decades ago and was initially formalized in the Basic Agreement. As I interpret the proceedings of autumn 2012, the parties actively mobilized resources to defend their understanding of the institution and the way it had traditionally been practised. They had earlier accepted changes, yet they did not accept these changes. Though perspectives based on new institutionalism and perspectives focusing power- and actors’ resource mobilization may be seen as competing alternatives, they can also be regarded as complementary. Both actors and structures have a potential role as driving forces for change, yet new measures may be hindered, or disappear, if they are actively opposed by “significant others” within the institution (Olberg & Trygstad, 2007, p. 139). Regarding this case, this seems to be what happened by the end of 2012.

Conclusion 

 153

8.5 Conclusion The long-term effects of the financial crisis and the following debt crisis in several European countries are yet to be seen. In the Norwegian context strong state finances were undoubtedly vital in order to respond to the sudden downturn in 2008/2009. Yet, it was also important that there existed a tradition for institutionalized tripartite cooperation and that schemes such as the temporary layoff institution were already in place and had been practised for many decades. Compared to Sweden, but also to Denmark, the coverage of social security contributions relating to temporary layoffs and short-time work was broader, and the social partners had greater influence on adjusting regulations. This chapter has explored the role of the social partners in the process of incrementally changing regulations of temporary layoffs. At the central level, in the initial phase of the financial crisis 2008/2009, the employer and trade union confederations cooperated successfully in pushing for relaxed regulations. The regulations covering the state’s economic contribution to the system of temporary layoffs were changed several times, generally in line with claims made by the central social partners. In 2012 the LO and NHO again joined forces and successfully pressed for changes; this time the issue was not the regulations per se but the more restrictive practices administered by NAV. New rounds of bipartite cooperation between the LO and the NHO took place in both late 2013 and winter 2014. These rounds also concerned temporary layoff rules and practices, and largely the parties’ claims were acknowledged by the minister of labour. The only significant difference seems to be that this time the third party was a conservative-right government. Shifting government coalitions do not seem to matter much as far as the temporary layoff institution is concerned. Nor is it something new that the LO and NHO join forces and mobilize to influence the temporary layoff regulations; this was done also in 2003-2004 (Nicolaisen & Lismoen, 2004). More generally the outline above has illustrated that a coalitional perspective is useful in order to explain both stability and change. The case is also one of many illustrations of shifting coalitions within the Norwegian cooperative model. Within institutional theory, it has been argued that the continuum models and dichotomous structural variables which the VoC literature often rely on make it difficult to analyze different combinations of, for instance, equality and liberalization or solidarity and coordination. Suggesting a political-coalitional approach, Thelen (2012, p. 16) draws on the strengths of power-resource theory (employer interests are conceived and articulated in a context in which the power and organization of labour are a key fact), labour market dualism theory (contemporary market trends

154 

 Regulating the Temporary Layoff Institution – Coalitions and Drift

complicate the unity on the labour side) and corporatism theory (high level of employer organization as a precondition for continued high levels of social solidarity).7 A significant feature of the temporary layoff institution is its capacity to be regulated “up” and “down”, illustrating, in an institutional perspective, gradual change through drift. More generally, this seems to be drift of a special kind – powerful actors regard the institution as “their” institution. They accept changes within certain limits, and there is room for discretion in the interpretation of regulations. Yet, they actively monitor changes in rules and practices, and from time to time they join forces in order to prevent regulations from drifting too far. In this sense bipartite coalitions within the tripartite cooperative model illustrate a source of stability.

7 Adding to the comment on the corporatism school, Thelen notes a need to “draw clearer distinction between the structure and the content of tripartite bargaining in order to move beyond the idea of corporatism as a static feature of countries, and to focus instead on ongoing renegotiation and contestation over the form and the functions of corporatist bargaining over time” (Thelen, 2012, p. 16).

References 

 155

References Andersen, S. K. (2009). Tackling job-losses – Varieties of European responses. Article presented at the fourth EU-China High Level Roundtable on Social Security, Stockholm 15-16 October 2009. Copenhagen: Faos. Berg, T. (2010). Hovedavtalen og den norske arbeidslivsmodellen. In T. Bergh (Ed.), Avtalt spill (pp. 13-35). Oslo: LO and NHO. Due, J., & Madsen, J. S. (2010). Septemberforliget 1899 og udviklingen af den danske aftalemodell. In T. Bergh (Ed.), Avtalt spill (pp. 315-339). Oslo: LO and NHO. Dølvik, J. E. (2013). Grunnpilarene i de nordiske modellene. Et tilbakeblikk på arbeidslivs- og velferdsregimenes utvikling. Delrapport 1, NordMod 2030. Oslo: Fafo. Eurofound Eiro (2009). Working time developments – 2008. Eurofound News, September 2009. Evju, S. (2007). Permittering og trygd: Trekk ved utviklingen. Arbeidsrett og arbeidsliv, 2, 135-184. Glassner, V., & Galgóczi, B. (2009). Plant-level responses to the crisis: Can jobs be saved by working less? ETUI Policy Brief. European Economic and Employment Policy, 1. Brussels: European Trade Union Institute. Glassner, V., & Keune, M. (2010). Collective bargaining responses to the economic crisis in Europe. ETUI Policy Brief. European Economic and Employment Policy, Issue 1. Hall, P. A. (2010). Historical institutionalism in rationalist and sociological perspective. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change. Ambiguity, agency, and power (pp. 204-223). Cambridge: Cambridge University Press. Hall, P. A., & Soskice, D. (2010). Introduction. In P. A. Hall & D. Soskice (Eds.), Varieties of Capitalism. The Institutional Foundations of Comparative Advantage (pp. 1-68). Oxford: Oxford University Press. Hansen, K., & Kvadsheim, H. (2008). Permitteringsordningen – en gjøkunge i NAV-systemet? Stavanger: IRIS. Hansen, K., & Kvadsheim, H. (2010). Bedrifters bruk av permitteringer. Søkelys på arbeidslivet, 1/2, 42-58. Hijzen, A., & Venn, D. (2011). The role of short-time work schemes during the 2008-09 recession. OECD Social, Employment and Migration Working Papers, No. 115, OECD Publishing. Jørgensen, H., & Schulze, M. (2010). Leaving the Nordic path? The role of Danish trade unions in the welfare reform process. Social Policy & Administration, 45, 206-219. Kullander, M., & Henrikson, J. (2010). Wage bargaining rounds for 2010 concludes. Eiro online. http://www.eurofound.europa.eu./eiro/2010/06/articles/se1006019i.htm. Leschke, J., & Watt, A. (2010). How do institutions affect the labour market adjustment to the economic crisis in different EU countries? Working Paper 2010:04. Brussels: European Trade Union Institute. Lundh, C. (2010). Saltsjöbadsavtalet och arbetsfreden. In T. Bergh (Ed.), Avtalt spill (pp. 293-313). Oslo: LO and NHO. Mahoney, J., & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change. Ambiguity, agency, power (pp. 1-37). Cambridge: Cambridge University Press. March, J. G., & Olsen, J. P. (1984). The new institutionalism: Organizational factors in political life. The American Political Science Review, 78, 734-749. Martin, C. J., & Swank, D. (2012). The political construction of business interests. Coordination, growth, and equality. Ambridge: Cambridge University Press. Nergaard, K. (2010). Temporary layoffs – saving jobs or delaying readjustments? Mutual learning programme 2010. Peer Review on France 13.09.2010. GHK Consulting and CERGE-EI. Nicolaisen, H., & Lismoen, H. (2004). Permittering. Konsekvenser av nye regler. Oslo: Fafo.

156 

 References

NOU 2013: 13 Lønnsdannelsen og utfordringer for norsk økonomi (Official Norwegian Report, chapter 3, Institutional relationships and the wage formation. Oslo: Ministry of Finance) Olberg, D. (2010). Nedgangskonjunkturen og arbeidstiden – permittering og redusert arbeidstid 2008/2009. Fafo-notat 2010:18. Olberg, D. (2013). Bruk av permitteringer 2012 – omfang og reguleringer. Fafo-notat 2013:08. Olberg, D., & Trygstad, S. C. (2007). Organisasjonsendring – om moter, myter og praksis. In J. E. Dølvik, T. Fløtten, G. Hernes & J. M. Hippe (Eds.), Hamskifte. Den norske modellen i endring (pp. 125-144). Oslo: Gyldendal. Opsal, V. (2007). Permittering, trygd og lønnsplikt. Arbeidsrett og arbeidsliv, 2, 185-302. Scott, W. R. (1995). Institutions and organizations. California: Sage. Streck, W. (1992). Social institutions and economic performance. Studies of industrial relations in advanced capitalist economies. London: Sage. Svalund, J., Casinowsky, G. B., Dølvik, J. E., Håkansson, K., Jarvensivu, A., Kervinen, H., Møberg, R. J., & Piirainen, T (2013). Stress-testing the Nordic models: Manufacturing labour adjustments during crisis. European Journal of Industrial Relations, 19, 183-200. Sørbø, J., & Bråthen, M. (2009). Permitteringer i en nedgangskonjunktur. Arbeid og velferd, 3, 34-38. Thelen, K. (2012). Varieties of capitalism: Trajectories of liberalization and the new politics of social solidarity. Max Weber Lecture Series MWP – LS 2012/03. European University Institute. Torp, H. (1987). Permitteringer i Norge – omfang og fordeling. In H. Indreberg & H. Torp (Eds.), Permitteringer og arbeidsløshetstrygd. En konferanserapport (pp. 10-18). Oslo: Institutt for samfunnsforskning.

Jørgen Svalund

9 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective1 Dismissal regulation constitutes a basic feature in the organization of labour markets. Dismissal regulations restrict the managerial capacity to dismiss employees to allow for downsizing (Regini, 2000, p. 16), and mainstream economics have often blamed such regulations for the low job growth in the European labour market (Pissarides, 2001). Thus, some propose that laxer employment protection regulation will increase employment and create more flexible labour markets. The employment protection legislation in the Nordic countries is, comparatively speaking, medium strict, according to the OECD employment protection legislation index (OECD, 2012b). The core theme of this chapter is how the dismissal regulation institution influences management choices during company-level crises within the Nordic countries. Do institutional practices follow strictly from the institutional rules or regulations themselves, or do actor relations influence the choices and practices during labour adjustment situations? Is it possible that the regulation of employment protection may lead to diverse outcomes, due to differences in competence requirements, the use of manpower, technologies and so on? Thus, the analysis provides insight into how institutional regulations may result in diverse outcomes, providing ground for incremental institutional change. Katzenstein (1985), among others, has highlighted that the Nordic labour market regimes enable adjustments during shifting economic circumstances. In his view the predictable and continuous ongoing relations between the social partners and the state enhance the degree of trust and institutional flexibility at all levels of the labour market. This is a system where preferences are traded off against one another and where the idea of the common good and a culture of compromise prevail over employer or employee interests (Katzenstein, 1985, p. 33). These compromises may be understood as an institutionalization of a certain distribution of goods and burdens (Elster, 1992; J. Mahoney & K. Thelen, 2010). Economic crisis and labour adjustment situations accentuate the need for flexibility and cooperation at the company level while potentially crystalizing conflicting interests and triggering mobilization of power resources. Thus, they provide the possibility of studying whether and how trust and long-lasting reciprocal relations influence institutional practices during times of institutional pressure. Combining laws and collective agreements, industrial relations are organized in a two-level system in the Nordic countries. Wages, productivity, working hours and

1 The empirical research was financed by the research project “Nordic models facing crisis” within the Norwegian Research Council research program VAM. © 2015 Jørgen Svalund This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

158 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

other working conditions are first bargained centrally at the national or industry levels. These centrally negotiated frames provide local unions with the right to consultation and participation regarding a range of workplace issues and a role in local collective bargaining (Falkum, 2008; Løken, Stokke, & Nergaard, 2013). As part of this system unions2 and employers are obliged to discuss alternatives to dismissals, such as temporary layoffs, reduced working hours, voluntary temporary or permanent pay cuts or other possible measures, as well as the extent of dismissals and their distribution (Svalund, 2013; Svalund et al., 2013). But the employer has the final say, and the company unions have no right to invoke industrial action in such instances (Stokke, 2008). J. Mahoney and K. Thelen (2010, p. 19) assert that two main properties influence the type of actions possible within institutional structures: firstly, the characteristics of the political context and whether or not it provides the opposing actors strong veto possibilities, and secondly, whether the targeted institution affords actors opportunities for exercising discretion in the interpretation or enforcement of a rule or regulation. During labour adjustments, the employer and the unions may discuss how the dismissal regulation should be understood or enforced. Furthermore, the relations between the actors involved in labour adjustments in the Nordic countries are often formed by both cooperation and opposition (Huzzard & Nilsson, 2004). Thus, the form of the social exchange relationship between the actors involved affects the outcomes of institutional rule enactment. Further, actors do, in many situations, face several alternative institutional choices. To avoid the burdensome consequences of dismissal regulation the actors may choose to adjust in ways regulated by alternative or complementary measures, choosing to use, for instance, severance pay to avoid dismissals distributed by the seniority rule (Dahl & Nesheim, 1998; Smith, Masi, van den Berg, & Smucker, 1995). It is therefore necessary to also focus on the relation between the targeted institutional measure and other institutions/institutional measures influencing how actors act towards, for instance, the regulation of dismissals (Hall & Soskice, 2001; Sels & Van Hootegem, 2001). The following presentation is based on a comparison of the rules and regulations of the dismissal institution and other relevant labour adjustment institutions in Denmark, Finland, Norway and Sweden, as well as 21 comparative case studies of companies adjusting labour within the newspaper industry and construction industry in Norway and the manufacturing industry within all these four countries (Svalund, 2013; Svalund et al., 2013).3 Placing the actors at the centre of the analysis provide

2 Union is used as a short term for trade union. 3 These are presented in Svalund et al. 2013 and Svalund 2013, but with a somewhat different theoretical focus.



Analytical Framework 

 159

insight into their available choices and offers detailed insight into the sources of institutional diversity and change. The chapter is divided into five sections. Firstly, I discuss the concept of institutions and how institutional diversity and change may be understood and explained. Secondly, the importance of understanding the dismissal institution as related to other institutions providing and regulating other labour adjustment possibilities, each with their own distributive outcomes, is discussed. Thirdly, I argue that broadening the understanding of the actors involved in institutional interpretation, enforcement and change is important for understanding how they relate to institutions and to understand how ambiguity, relations and power lead to diverse institutional outcomes within different industries and in relation to differences in the institutional measures available within different countries. In the fourth section I show, using examples from comparative case studies, how the set of measures facing the actors and the established social exchange relationships influence institutional reproduction and flexibility. Finally, the relevance of the perspective presented is discussed.

9.1 Analytical Framework 9.1.1 Institutions, Agency and Outcomes Scott points out that “Institutions are comprised of regulative, normative and cultural cognitive elements that, together with associated activities and resources, provide stability and meaning to social life” (Scott, 2008, p. 48). What defines institutions is not that actors agree with and follow the norm, but rather that it is obligatory to follow it, the rules and regulations may be enforced by a third party (J. Mahoney & K. Thelen, 2010; Stinchcombe, 1968, pp. 154, 188; Streeck & Thelen, 2005). Institutions reduce uncertainty, increase the level of trust between actors and create effective frameworks for decisions (Crouch, 1999, p. chapter 2; Streeck & Thelen, 2005, p. 9; Svalund & Kervinen, 2013). By creating predictability and social order, institutions not only limit the number of choices available, but they also establish criteria and resources that individual and collective actors use to make their choice of action (Giddens, 1984; Immergut, 1998, p. 26; Wailes, Ramia, & Lansbury, 2003). Still, there is usually room for actor discretion in the way rules and regulations are enacted or enforced.

9.1.2 Labour Adjustment Strategies in the Face of Institutional Regulations Rule makers such as company actors facing crisis and a need for labour adjustments relate to a set of labour adjustment measures, each with their own rules and regulations. In practice, most managers meet numerous and sometimes incompatible demands from a range of external actors (Pfeffer & Salancik, 2003), such as owners/

160 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

stockholders, the administration in their parent company, employees and unions and the local community, as well as from rules and regulations governing labour adjustment measures. Management must balance these demands, acting in ways that to differing degrees imply resistance towards the institutional rules and regulations. Hence, while institutional diversity and change may strategically be sought by change agents, it may also be a by-product of events and decisions not aimed at institutional change per se. Oliver (1991, p. 152) presents a typology with five types of strategies that company actors may enact when facing institutional demands, varying from passively following the rules to active resistance (Oliver, 1991, p. 151). Actors may, firstly, use the strategy of acquiesces during dismissal selection, following taken-for-granted norms, mimicking institutional models and trying to obey regulations to the letter. Secondly, they may use a strategy of compromise, balancing the expectations of multiple constituents, trying to “bend the rules” and negotiate with institutional stakeholders. Thirdly, they may try to avoid the regulations altogether by changing goals, activities or domains by, for instance, shifting from the use of dismissals to the use of voluntary quits by way of severance pay. Fourthly, the company actors may try to defy the institutional rules by ignoring or contesting explicit norms and values or by assaulting the sources of institutional pressure. Finally, the companies may choose to try to manipulate the institutions by way of co-opting, influence or control those influencing the institutional rules. As mentioned, J. Mahoney and K. Thelen (2010, p. 19) assert that the political context and whether or not it provides the opposing actors strong veto possibilities, and whether the targeted institution affords actors opportunities for exercising discretion in the interpretation or enforcement of a rule or regulation, influence the type of strategies and actions possible when actors face different institutions. While institutional practices may change over time, for rule takers, the political context, as well as the level of opportunities for discretion in enactment, opens up for diversity between, for instance, industries at one point in time. Thus, changes in industry structure (Bell, 1976) may increase the importance of sectors with different institutional practices, leading to incremental macro-level institutional changes.

9.1.3 Power and Compromises Within Long-Term Reciprocal Relationships Labour adjustment negotiations are embedded in long-lasting relationships, where order and long-term reciprocity are necessary to maintain good, cooperative relationships (Granovetter, 1985). In contrast to, for instance, Germany, employers in the Nordic countries are given the final decision-making authority on labour adjustment decisions (Sippola, 2012; Svalund & Kervinen, 2013). The political context does not provide the unions with the opportunity to veto, for instance, a decision to dismiss a certain number of employees. Still, violating established normative

Data 

 161

and cultural expectations, and disregarding advice given by the unions during consultations, may have detrimental effects on the employer–employee relationship. Hence, trusting, long-lasting relationships provide opportunities for other types of social exchanges than non-trusting relationships (Crouch, 1993). As this relationship does not only relate to the potential dismissal situation at hand, but also to future wage bargaining, productivity, product development and so forth, the concept of the social contract, applied within strategic choice theory, points to how social relations might influence the significance of the political context (Walton & McKersie, 1965; Walton, McKersie, & Cutcher-Gershenfeld, 1994). The term social contract refers to the shared understandings between management and unions about the meaning of cooperation, the “rules of the game” in bargaining and cooperation (Walton et al., 1994, pp. 3-11). The existence of a social contract, a common history and shared understandings between management and unions may influence the labour adjustment measures chosen and the process of negotiation.4 Hence, the actors’ bargaining power also rests on the fact that they are bound to interact in the future and must judge whether use of power may have detrimental long-term effects on their relationship (Walton et al., 1994, p. 43).

9.2 Data The financial crisis of 2008, with accompanying demand and revenue reductions across countries and industries, offered a unique opportunity to study such issues within several industries and countries facing different constraints and regulations. Case studies were conducted within three construction companies, six manufacturing companies and three newspapers in Norway. These three industries are highly unionized among their core employees and experienced a substantial downswing due to the financial crisis. Manufacturing, a cornerstone of the Nordic labour market model, could then be compared with construction – an industry with long-term projects and fragmented and dispersed product organization – and newspapers, producing intangible products with highly skilled white-collar employees. Further, case studies of manufacturing companies were also conducted in Denmark (three companies), Finland (three companies) and in Sweden (three companies), enabling a comparison of companies within a single industry facing different regulations and measures. Companies with more than 50 employees prior to the adjustments were selected, giving insight into choices, norms and perceptions of justice in regard to

4 “Negotiation(s)” is in this chapter understood as “exchange(s)”, and should not be understood as necessarily formal negotiations. The unions have only formal consultation rights, but other power resources may turn these into negotiation processes of sort.

162 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

distributional effects not available in smaller companies. Data were collected through interviews with representatives of site management (HR directors and/or production/ division managers) and of the main blue- and white-collar unions. The interviews focused on the consequences of the crisis on the company, how the adjustments were handled and whether and how management and unions related to each other and the rules and regulations they were facing. Interview data were supplemented by annual reports, business data and management–union agreements. A common interview guide was used in the fieldwork, and to ensure data quality all interviews in Denmark, Norway and Sweden were summarized and emailed to the interviewees for comments. For assessing the institutional norms and regulations, relevant statutory laws and collective agreements regarding labour adjustments was consulted

9.3 Labour Adjustments in the Nordic Countries: Institutional Characteristics and Alternatives There are three main types of institutional measures which are nationally regulated, either through law or collective agreements, and I present each below.

9.3.1 Dismissal Regulation The 1998 EU directive on mass redundancies, applying to all four countries, states that employers must consult with employee representatives before making a decision involving redundancies of 10 or more employees. Discriminatory decisions are forbidden in all countries, but there are other regulatory differences. In Denmark the Basic Agreement between Landsorganisationen i Danmark (LO) and Dansk Arbejdsgiverforening (DA) states that dismissals must be based on reasonable grounds related to the employee or the company, but criteria for selection are not defined. The same pertains to Finnish legislation, while the Dismissal Protection Agreement between Suomen Ammattiliittojen Keskusjärjestö (SAK) and Elinkeinoelämän Keskusliitto (EK) contains guiding principles emphasizing the importance of retaining skilled employees and securing those with reduced work ability. Lack of specification also pertains to the laws in Norway, but the Basic Agreement between Landsorganisasjonen i Norge (LO) and Næringslivets Hovedorganisasjon (NHO) specifies a seniority rule, which permits deviation from this principle with due reason. The last-in-first-out principle is mandatory under Swedish law as long as the remaining employees have qualifications to carry out the work.

Labour Adjustments in the Nordic Countries: Institutional Characteristics and Alternatives 

 163

9.3.2 Regulation of Temporary Layoffs There are state-funded temporary layoff or work-sharing schemes in Finland, Norway and Denmark (Olberg, this volume). The schemes have many similarities, most importantly payment of unemployment benefits to laid-off employees. However, the state-funded UB commence only after an initial period of layoff or short-time working; until then, the cost is borne by the worker, the employer or both. There is also a minimum proportion of time off work before payment is available: 40 per cent in Finland and 50 per cent in Norway, while in Denmark the employee has to be off work at least two days per week or one week off followed by one week at work. In Denmark, temporary layoffs have to be shared evenly between employees in at least a whole production unit, and there could be no redundancies within the unit during worksharing. By contrast, in Sweden there are no such state-subsidized schemes. In the aftermath of the 2008 crisis, the main employer and employee organizations within manufacturing in Sweden were able to agree on a “crisis agreement”, providing a sort of temporary layoff mechanism, without state subsidizing (Svalund et al., 2013).5

9.3.3 Unemployment Benefits and Early Retirement Systems During labour adjustments the actors may take the unemployment benefit and early retirement schemes into account. There are only minor differences in the generosity of the insurance schemes in the Nordic countries (OECD, 2012a). At the time of the crisis, Denmark had an allowance (efterløn) enabling workers to retire starting at the age of 60; in Finland individuals turning 60 before the 500-day period of unemployment benefits was over could move directly into old-age pension; in Norway a statesubsidized collectively agreed upon supplementary pension scheme allowed those covered to retire at the age of 62. In Sweden there was no such publicly subsidized early retirement system, and the normal retirement age was 65. These institutions complement each other, and present the company actors with a set of choices. Simultaneously, the dismissals selection rules provide more possibilities for employer discretion in Denmark and Finland than in Norway and especially Sweden. In Sweden temporary layoffs are unavailable or very costly to the employers and employees; whereas Denmark and Finland have lenient dismissal selection rules, these countries also have more lenient early retirement regulations, providing more employer choice.

5 There have been several minor adjustments in these rules in the last few years for more on this see Chapter 8 (Olberg) in this book.

164 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

9.4 Power, Cooperation and Institutions within Three Industries in Norway Studying companies with different production methods, competence requirements and cooperative relations provides a way to compare how the same set of institutions and measures enable and constrain rule takers and their power resources differently in different situations. The institutions structuring labour adjustments may provide possibilities for different types of strategies in different types of situations, thus comparing adjustments in companies in three different industries highlights how the same regulation may lead to very diverse institutional outcomes. Table 9.1: Types of measure used by company and industry Category of measure

1. Reduction of external employees

Subtype

Stop/reduction in the use of consultants and agency workers

2. Temporary adjustments 1. Flexible working time in working hours arrangements, collective agreements 2. Temporary layoffs

Company Manufacturing Construction

Newspapers

M2, M3, M4, M6*

All

All

All

 

M2

All

3.Training measures 4. Cuts in working time and pay by CA 3. Internal re-deployment*  

NP2 M1, M2, M3, M5

 

NP1

M2, M4, M5, M6

C2, C3

NP1, NP2

4. Permanent changes in pay systems

 

5. Exits and early retirement 1. Voluntary resignments – attrition with severance pay

 

2. Early retirement incentives 6. Dismissal/involuntary quits

C1

All

(C1)

All

1. Phasing out temporary employees

M2, M4, M6

 

NP1, NP2

2. Dismissals

All

All

NP3

Table 1 is a modified version of Svalund (2013, Table 3). M1 stands for manufacturing company nr 1, C2 for construction company nr 2, etc.



Power, Cooperation and Institutions within Three Industries in Norway 

 165

Production contingences and the way the crisis hit the companies influenced the choice of measures (Table 1) in specific ways, leaving most room for agency and union power in the newspaper industry and least in construction. It also influenced the cooperation between management and unions and the power resources of unions during these processes.

9.4.1 Manufacturing The financial and economic turmoil of mid-2008 resulted in a drop in demand that was quick and sudden in the manufacturing plants. Just in time (JIT) production methods swiftly made demand reductions evident on the plant floor, as production was attuned to incoming orders. Further, the uncertainty of the evolving economic situation made it difficult to create long-term adjustment plans, making combinations of temporary measures and dismissals, sometimes with redeployments, more suitable. The overriding goal for both management and unions was to rescue as many jobs as possible at the plant and retain the competences and skills necessary to be productive in the future. The unions wanted to move the burdens on to external workers first, but the size of this group was very limited and of no real effect. Both actors realized that there was limited room for distributive win-lose power games. Handling the situation effectively could reduce the number of dismissals, while conflicts could increase the crisis and the number of dismissals at the plant. Further, both parties knew they could not offer much in terms of severance pay or keeping workers without tasks employed, as there were limited economic resources available at the plant level. The speed of the crisis and the fact that temporary layoffs initially required that employees were laid off at least 40 per cent in a maximum of 30 weeks made both management and unions in several companies interested in finding other short-term solutions. Therefore, the unions engaged in cooperative efforts enabling win-win solutions rather than forcing the employer into win-lose power games. In the first manufacturing company (M1) the union negotiated a local “crisis” agreement where wages were cut 10 per cent while working time was cut 20 per cent, sharing the burden between employer and employees. Similar measures were created in three other companies, and the unions were essential in building consent among the employees (Table 1). Cooperation saved jobs, enabling solutions where dismissals were avoided and at the same time provided the unions further leverage and goodwill in later negotiations. Even though other measures were used, all the companies had to dismiss employees. The interpretation of the rules for distributing dismissals was pivotal in the negotiations. The unions wanted a strict application of the seniority principle, while management in some cases wanted exceptions on account of competence. While there were discussions about what constituted “a difference in seniority” in some of the

166 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

companies, the unions were able to get their opinion through, applying the criteria rather strictly. In the end, both management and unions were usually happy with the principle, as it saved skills, was judged to be fair and enabled an effective, simple selection process. Also, while the unions could not veto the use of other criteria, they could use their informal and formal day-to-day cooperation as leverage, making management accept their definition. As such, cooperation between management and unions on issues such as the creation of short-term adjustment measures, on competence, productivity and wage provided the union with leverage that they could use as an informal local veto power, reducing management’s possibilities to interpret the dismissal regulation loosely. Further, none of the companies used general severance packages or early retirement offers since they followed a strategy of acquiescence. Following a strategy where the dismissal process was done strictly according to laws and regulation, severance pay was just not perceived to be necessary by management. Also, the unions did not try to get such packages, or they did not prioritize it against other claims during the negotiations. Another way of avoiding dismissals and the dismissal rules is early retirement. Most of the companies had been working to keep and attract senior employees over the previous years, and as such, they found the use of early retirement packages to be normatively wrong. Thus, the manufacturing plants ended up applying a rather strict interpretation of the dismissal rules while also creating local flexibility through locally agreed upon measures.

9.4.2 Construction The drop in demand had a slower impact in construction as the need for man-hours on ongoing projects remained the same, while the companies did not have enough new projects at hand. Initially the companies aimed at keeping more of their future projects in-house, limiting the use of sub-contractors and agency workers. The unions’ main aim was to keep as many workers employed as possible, preferring general cost-cutting and stops in the use of agency workers and temporary contracts over temporary layoffs and temporary layoffs over dismissals. While the employers shared the interest in protecting employees, especially those with valuable skills, both parties realized that dismissals would be necessary, in combination with other measures (Table 1). The use of measures and the interpretation of dismissal regulations in the construction companies can be understood as being based on compromise and avoidance, varying somewhat between the three cases. The application and distribution of dismissals and temporary layoffs were important. The unions wanted the selection of dismissals to be done according to seniority within each occupation. Since seniority and productivity were not necessarily correlated, the employers were more concerned about keeping the skilled, motivated and productive employees than



Power, Cooperation and Institutions within Three Industries in Norway 

 167

selecting mechanically according to seniority. Letting some of the less productive employees go would be preferable in their view. Facing conflicting interest, defining the operational application of the dismissal selection regulations was important. Management used court rulings and law interpretations to back up that the difference in seniority between employees had to be substantial, up to several years of difference in seniority. In C1 the blue-collar union disagreed with management. Still, many employees were laid off, and despite union resistance the employees were offered a small severance package or early retirement package if they quit voluntarily. Some were offered to quit immediately while keeping their wages through the notice period. In addition the company had a “wage guarantee”. Those lacking new employment before the notice period ended were offered a continuation of their wages for some time, usually one to three months, based on seniority and social considerations. As such, management used a strategy of avoidance, moving the manpower reductions from a situation where they were influenced by the dismissal regulation to “voluntary resignations”. In C3 the perception of “usefulness” was stretched further than in C1, implying both specific craft competence and motivational aspects. Early on in the adjustment process there was a conflict between the local management and the bluecollar union regarding the selection criteria, with management wanting usefulness/ competence to be the primary criterion, before seniority. The manager was able to convince the local blue-collar union to use usefulness/competence as the primary criterion, using the piece-rate system as an important argument. It was argued that saving what management considered to be the “best” employees could potentially improve the work climate and increase the piece-rate reward for the “survivors” in the future, while using a strict seniority criterion could impair productivity and hence wages and job security of all remaining employees. The result was that “a difference” in length of seniority was widely defined in many instances, making competence the de facto criterion. Also, by using a wide definition of “competence”, the dismissal selection was in effect mostly up to the negotiations between management and unions to settle. Hence, the company avoided following the dismissal regulation to the letter. In C2 a relatively high number of white-collar workers did quit, either by choice or by being dismissed, whereas only one blue-collar worker was dismissed. While the application of dismissals was important for the company-level actors, they also used different strategies to reduce the number of dismissals. All unions made sure that internal redeployments were used and that the use of agency workers was terminated. Further, the blue-collar unions in C2 and C3 demanded that workers were hired out to other construction companies in the region, in accordance with clauses in the central collective agreement. As such, facing somewhat strict dismissal rules, management preferred several alternative measures to avoid the dismissal rules. These construction companies partly avoided following the dismissal rules to the letter. This can be attributed to the fact that the blue-collar unions had few resources to rely on as they were facing competition from subcontractors and agency workers Their members were less indispensable in the production process than was the case

168 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

in manufacturing, and their role on the actual building sites regarding productivity and other work-related issues was more limited.

9.4.3 Newspapers The crisis hit the newspaper industry almost instantly, as they all experienced a sudden drop in advertisements during the fall of 2008. The newspapers lost revenues while the need for man-hours to produce the daily news remained more or less the same. The changes in the newspapers were, compared to manufacturing and construction, much more oriented towards reorganizing activities to save labour input and wage costs. As such, the crisis accentuated ongoing reorganization aimed at tackling the structural changes already under way in the news industry. Several measures were used to adjust the use of labour: The companies reduced their use of temporary employees and freelancers, and as they used such manpower extensively prior to the crisis, these cuts had substantial impact. In NP1 the typographer union instigated an employment pact where half the agreed wage increase in 2009 was returned to the employer in order to save two typographers who otherwise would have to quit because of low seniority. Following this, all quits among the typographers were early retirements, the preferred option for the group. Finally, the companies chose to avoid the dismissal regulations by using severance pay and early retirement packages, thus avoiding the distribution of dismissal by seniority. While the unions in manufacturing and construction wanted the seniority rule to be applied, management wanted to avoid such a distribution of dismissals within the newspapers as it did not fit the company-specific competences. Within the media industry, and in these specific companies, former downsizing processes had resulted in the use of severance packages and early retirements, often rather generous ones. The union wanted general packages that everyone could apply for; they wanted the severance pay but to avoid that management could pick those management wanted to leave. More use of internal flexibility measures, reorganizations and changes in production processes made the unions more central than in manufacturing and especially construction, and resulted in tough negotiations and trade-offs between management and unions. The fact that the managers wanted to avoid the seniority principle meant that the journalist unions could use the regulations in trade-offs with management, demanding severance pay and “voluntary” resignations as an alternative if they were to contribute to fast, flexible and legitimate work-time cuts and reorganizations. The need for internal reorganization in the newspaper rather than cuts in labour hours made cooperation with the strong journalist occupation and their union all the more important. Further, former “victories” in the form of severance pay during previous labour adjustments had produced path-dependencies, making it hard for managers to argue against repeating it. The unions were aware of the power



Power, Cooperation and Institutions Within Manufacturing in the Nordic Countries 

 169

they possessed in the negotiations. The NJ representatives had backing from their members, and stressed both their traditional union power through good finances, that their members skills were essential for the development of the company in the future and that a good cooperative climate was essential for the future adjustment and production processes. A bad management– union relationship could influence the daily work spirit. Journalistic work is, compared to more repetitive work tasks in factories and construction sites, difficult to monitor, and editors and journalists cooperate closely in the search for and presentation of stories. Management in these companies acknowledged that the unions were essential for employee support in the adjustment processes and for further development of the company. As such, the need for cooperative relationships influenced how the dismissals were realized.

9.5 Power, Cooperation and Institutions Within Manufacturing in the Nordic Countries After comparing adjustments within manufacturing, construction and the newspaper industry in Norway, I discuss differences between adjustment processes in manufacturing in Denmark, Finland and Sweden. Comparing company level labour adjustment situations in a single industry within different national regulatory systems provides a basis for understanding the relevance of institutional diversity for cooperation and outcomes, and especially how variations in rules and available adjustment measures influence choices within the same industry. The choice of adjustment measures was influenced both by the perceived depth of the crisis and the institutional environment. Many companies needed to adjust labour and production capacity immediately. When the crisis broke out, the prime concern of unions was to shield their own members from loss of jobs and pay, while management was looking for ways to save costs. As a first line of defence, unions in many companies thus called for, and usually won, management accept for reduction of external labour in the plants, such as agency workers and consultants (Table 2). The case studies show that national differences in the regulation of dismissal selection criteria and in the availability of short-term work (temporary layoffs) schemes provided variations in the need for cooperative solutions and in the actors’ control over issues of interest to their opposing social partner.

170 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

Table 9.2: Types of measure by company and country Category of measure

Subtype

Manufacturing company DK

FIN

NO

SWE

 

FIN2

NO2 NO3 NO4 NO6

SWE1

DK2

All

NO2

All

DK2 DK3 All

All

 

DK2

FIN1 FIN2

 

 

 

3. Internal re-deployment***  

All

 FIN2

4. Permanent changes in pay systems

DK2 DK3  

 

 

5. Exits and early retirement 1. Voluntary resignments with – attrition severance pay 2. Early retirement incentives

All

 

 

All

DK1 DK2  

 

All

6. Dismissal/involuntary quits

1. Phasing out temporary employees

 

2. Dismissals

All

1. Reduction of external employees

Stop/reduction in the use of consultants and agency workers

2. Temporary adjustments in 1. Flexible working time working hours arrangements, collective agreements 2. Temporary layoffs/work sharing 3.Training measures

4. Cuts in working time and pay by CA

FIN2

SWE1* SWE2 SWE3* NO1 NO2 SWE1 NO3 NO5 SWE2 NO2 NO4 SWE1 NO6 SWE2

NO2 NO4 NO6 FIN2 FIN3 All

SWE1 SWE3 All**

*In SWE1 training measures were funded by the operating budget. In SWE3 an externally funded project covered the costs related to the training, while the wage costs were underwritten by the company. ** In SWE2 only a few white-collar employees were dismissed. ***Included hiring out. Table 2 is a modified version of Table 1, p. 191 in Svalund et al. (2013).

9.5.1 Denmark: Dismissals as an Employer Prerogative – Employer Choices without Union Interference The adjustments in DK3 involved fairly extensive cooperation between management and unions in an effort to find mutually acceptable ways to adjust working hours and labour costs while protecting employment security. In the two other companies



Power, Cooperation and Institutions Within Manufacturing in the Nordic Countries 

 171

consultation was at arms’ length (Walton et al., 1994). The unions were consulted late in the adjustment process, and their involvement was kept to a minimum, mostly as one-way formal information. The companies did to a limited degree use work-sharing schemes. A wish to retain general and firm-specific skills made management in two of the companies prefer work-sharing instead of dismissals for some employees (Table 2). Further, management was able to win union acceptance for reorganizing bonus systems in order to lower wage costs permanently. In DK2 management obtained substantial savings on pay, implementing a new pay and bonus system for blue-collar workers, while in DK3 a planned pay rise was converted into a bonus agreement. Overall, regular dismissals were the most important means of adjustment in all companies. While the employers and unions had a long-lasting reciprocal relationship in these companies, dismissal selection had been regarded as an employer prerogative for many years. As such, the employers and unions did not expect union involvement or influence over this matter. Therefore, with lenient rules and clear expectations, both actors chose institutional acquiescence, following norms and expectations in these situations. The employers’ criteria for distribution of dismissals were based on their perceptions of the future need for skills and capabilities; in DK2 employee motivation and sickness absence were also important considerations. The unions’ normative expectations of not being involved meant that their lack of involvement did not influence the social contract. While the unions did not expect it, the employers in DK2 and DK3 asked the union to forward their views, for instance, regarding social considerations when distributing the dismissals. In addition to dismissals, severance pay was to some extent also applied to promote employee loyalty, calm union discontent and provide incentives to keep up productivity while working through the notice period. Finally, Denmark has been known as holding a norm of early exit from the labour market (Hult & Edlund, 2008). A sizeable share (27 per cent) of the manpower reductions in DK3 was obtained through early retirements. As the union in DK3 aimed at avoiding dismissals, early retirements or severance pay was viewed as a better option, leaving the individual employees with a choice and some financial returns.

9.5.2 Finland: High Level of Employer Discretion during Dismissals Combined with Limited Veto Possibilities The adjustments involved some cooperation between management and unions in FIN1 and FIN2, where management formally consulted the unions to find mutually acceptable ways to adjust labour. The unions in FIN3 were involved late, and management kept their involvement to a minimum, mostly restricted to one-way formal information.

172 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

Finnish manufacturing companies have extended opportunities for temporary layoffs, as well as company-specific work-sharing schemes. This was used extensively to postpone and possibly avoid dismissals, supported by both sides in the plants. Still, regular dismissals were the most decisive means of adjustment in all companies but FIN1. The selection of whom to dismiss and retain was largely a management prerogative. The selection of dismissals in FIN2 and FIN3 was distributed with emphasis on job descriptions and skills and less on seniority. The unions were consulted but felt they had little influence regarding the number, selection and distribution of dismissals. In one of the companies with less trustful relations, the dismissals were perceived as an illegitimate “cleansing operation” (FIN3) by the union, while the CEO explained that the crisis created a perfect opportunity to remove unsuitable employees. The Finnish manufacturing companies faced a situation where the dismissal selection rules were rather loose and unclear, putting weight on a range of factors not possible to measure or veto and a temporary layoff arrangement that was cheap and easy to use. The combined effect was that there was little need for discussion, and acquiescence to the rules was perceived by both unions and management as the only logical choice. Further, the unions also had limited influence in other areas outside the specific dismissal rules, as the cooperative relations with management about production development, competence and wages were limited. Thus, they could not build on and use their cooperative relationship to persuade the employer to follow their wishes. Hence, institutional differences beyond those directly relevant to the dismissal processes also influenced the bargaining power of the unions.

9.5.3 Sweden: Lack of Temporary Layoff Mechanism Combined with Strict Dismissal Rules Laid Ground for Union Power In Sweden the adjustments involved extensive cooperation between management and unions trying to find mutually acceptable ways to adjust labour. Regular dismissals were the most widespread means of adjustment in all companies but SWE2. The overriding rule for dismissal of blue-collar workers was seniority, and deviations from the seniority rules needed to be justified and negotiated with the blue-collar union. To achieve immediate adjustments, companies initially drew on existing agreements regarding flexible working time (Table 2). Such agreements enabled hourly adjustments of around 10-15 per cent within a year. The actors in several instances re-negotiated their agreements or struck entirely new deals. This was usually instigated by the unions, which wanted to distribute the losses of work and income evenly while saving jobs. Further, the partners in SWE1 agreed on 20 “stopdays”, when production was closed down and workers were paid 85 per cent of their normal wage. Similar measures were seen in SWE2. Thus, the local partners negotiated measures in line with the central crisis agreement, even when not being



Discussion and Conclusion 

 173

covered by it. To complement the drive towards flexible working time adjustment, a training scheme (SWE1) and earmarked funds for education and in-service training (SWE2) were used to retain employees in the company. Faced with the diving market, management was forced to negotiate with the blue-collar union over how to save costs and skills by developing a scheme for internal working time flexibility that, in effect, served as a private functional equivalent to the publicly financed temporary layoff schemes known elsewhere. Instead of passing the costs on to the state, they were shared between the workers and the company. Finally, the wage agreements were delayed and moderate. The Swedish companies did not have any kind of state-sponsored buffers for labour hoarding or flexibility. Both management and unions would have preferred to have access to temporary layoffs. The absence of a temporary layoff scheme put greater pressure on the actors to negotiate mutually acceptable (win-win) alternative solutions. Thus, searching for alternatives, the employers needed to make compromises with the union. The combined effect of tight, long-term cooperative relationships between unions and management and sudden crisis while having a rigid, mechanical dismissal rule based on a last-in-first-out principle made other available labour adjustment methods all the more important. In the Swedish companies we find the clearest signs of give-and-take cooperation. The blue-collar unions would accept cuts only in working time and pay losses on the condition that there would be no dismissals within a certain period of time (SWE1, SWE2), that blue- and white-collar employees would be subject to fair distribution of the savings (SWE1) and that early retirement and redeployment would be offered (SWE3). Further, in these companies, voluntary exits with severance pay were frequently used as part of trade-offs in negotiating crisis packages. The unions could, for instance, demand such offers, especially for elder employees, in “return” for accepting dismissals. Early retirements or severance pay was viewed as a better option, avoiding dismissals and leaving the individual employees with a choice and some financial returns.

9.6 Discussion and Conclusion This chapter began by discussing dismissal regulation and how such rules have been attacked for creating rigid labour markets. While the Nordic labour market does not have the strictest rules on these issues in Europe, one aim of this chapter has been to investigate the flexibility of such rules, and whether and how their application is influenced by the political context, the local cooperative relations and the actor discretion allowed by the rules, as well as how the existence of alternative measures influences the political context and the choices made during labour adjustments. The Nordic manufacturing cases showed that national differences in the regulation of dismissal selection criteria and in the availability of short-term work

174 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

(temporary layoffs) schemes provided variations in the political context, in the need for cooperative solutions and in the actors’ control over issues of interest to their opposing social partner. Even though the temporary layoff mechanism in all three countries did not allow for the use of discretion, applying it saved jobs and skill investments and was therefore preferred by both employers and unions. The combined effect of tight, long-term cooperative relationship between unions and management, and the possibility to deviate from the seniority rule if due cause in Norway, made it possible for the unions to behave flexibly and with strength and to expect flexibility in negotiations on other issues in the future. In Sweden, the absence of a temporary layoff scheme put greater pressure on the actors to negotiate mutually acceptable (win-win) alternative solutions, thus enabling the new institutional mechanisms the Swedish employers needed to make compromises with the union. Without a temporary layoff arrangement, the employers and unions in Swedish manufacturing engaged, at both the company and industry levels, in tough discussions and trade-offs not necessary in Denmark and Finland, who had more lenient dismissal rules and temporary layoff arrangements. Hence, the combined effect of rigid dismissal rules and the absence of a temporary layoff scheme bred grounds for cooperation and union power, using discretion while operating within institutions. Hence, the set of alternative or complementary institutional mechanisms influenced the power of the unions, whether there were conflicts of interest between the employer and unions or not, and the strategies and choices made. Thus, while veto possibilities are important in some contexts, when studying variations in practices the available alternative strategies possible to deploy is as important. As witnessed in Sweden, facing company-level crisis and no publicly funded temporary layoff measure, the company actors articulated the need for such a measure to the sectoral and central actors. Building on a long-term trusting reciprocal relationship, the parties were able to create a new institutional crisis agreement, providing numerical company-level flexibility. While doing so, the institutional foundation of new measures and new institutional rules was contingent upon a strong reciprocal relationship. As such, the existence of strong cooperative relations and complementary institutions provided a basis for institutional stability, at least in the short run. Being more involved, the local unions were able through cooperative endeavours to provide new win-win solutions such as local crisis agreements on working time. While the tighter cooperation between management and unions in Norway and Sweden led to compromises and provided union influence, the Danish unions were mostly only informed about the dismissal selection process or the use of short-time work. As also shown by Hagen (1995) in the public sector in Denmark, unions, even though less involved that the unions in Norway and Sweden, found it natural that the dismissal selection was an employer prerogative. Hence, while regulative institutions influence whether the unions are involved, what Scott (2008) refers to as normative and cultural cognitive institutional elements, expectations of not being involved mean that a lack of influence does not necessarily influence the social contract.



Discussion and Conclusion 

 175

Comparing construction, manufacturing and the newspaper industry within a single country (Norway), the case studies clearly show that market conditions, production contingencies and work organization influence the possible choice of measures during the labour adjustment process and hence the possibilities for agency and power. This being so, these differences in production contingencies and skill needs influenced the cooperative relations and the degree of occupational union strength and control and thus the institutional processes and its outcomes. Management and unions in the manufacturing companies making products that were not large and time demanding had to cooperate to create possibilities otherwise not possible, agreeing on work time cuts and saving jobs and skills, which benefited both parties. There was less room for distributive win-lose power games compared to the newspapers, since they had much less economic resources available. In construction, the blue-collar unions had even fewer resources to rely on. Facing competition from subcontractors and agency workers, the main blue-collar occupational group had no closure on their work tasks. The union could not, compared to the journalist unions in particular, rely on their members’ indispensability in the production process. In newspapers, the need for the reorganization of production and work organization, and saving through work-time cuts, put the unions in a more central negotiation position than in manufacturing and construction. The strategies and institutional outcomes in these cases are firmly anchored in the social contract between management and unions at the company level. The influence of the unions and their use of power was reliant on the way production and the organization of work influence the need for cooperation. The unions, when possible, based their power on skills, cooperation and reciprocal exchanges over time. The depth of the day-to-day cooperation between unions and employers was important for union power during labour adjustments. Employers in manufacturing and the newspapers had invested in building cooperative relations with the unions for years, and wanted these to continue, while there were less cooperative traditions to rely on and uphold in construction. The less reciprocal relationship with the employers in construction reduced their bargaining power through the social contract, reducing the possibilities of management using compromise as a strategy. Further, the dismissal rules impacted the dismissal processes even when not applied. While national measures may fit some industries more than others, industrylevel path-dependencies and norms of appropriateness (March & Olsen, 1984) structured the labour adjustments and the impact of national regulations. While dismissal selection through seniority fit the company-specific competences within manufacturing fairly well, this was not the case within the newspapers. The journalist unions could therefore use the regulations in trade-offs with management, demanding severance pay and “voluntary” resignations as an alternative to dismissals by seniority if they were to contribute to fast, flexible and legitimate work-time cuts and reorganizations. In the newspapers, the use of severance pay during previous labour adjustments produce path-dependencies, expectations. Breaking these expectations

176 

 Cooperation and Power in Labour Adjustment Choices: A Nordic Perspective

could undermine the social contract and the cooperative relations with the unions. As such, the characteristics of the local political context and the local power relations led to the use of both strategies of institutional avoidance and acquiescence. While a long term reciprocal relation seemed to be of less importance within construction, the employers in manufacturing and newspapers were eager to hold on to a good relationship. Thus, veto power based on regulations was not the only or prime motive when adjusting labour, suggesting that the employment protection regulations in the Nordic countries are rather docile and that their application is dependent on the combined effect of employee power and the productive need facing the companies. Avoiding the effects of the seniority principle in the newspaper industry in Norway was important both to avoid the outcomes of the rule, for instance removing some of the young, skilled employees, but also since severance pay was an established norm within the industry. Thus, not paying severance pay could threaten the social contract between the local parties, damaging the relations necessary to develop the company further. As such, former institutionalized practices and the strength of the social contract together created institutional stability. This chapter has provided insight into the significance of complementary institutions for explaining institutional stability and institutional change from below. Comparing labour adjustments within different industries, I have shown that the impact of regulations changes due to shifts in the environment, and how, conditioned by the employers’ need for a certain quality of the social contract, the employers may bend the interpretation of the rules, pushing a certain compromise, a certain interpretation of the regulations forward. The typology of J. Mahoney and K. A. Thelen (2010), pointing to the characteristics of the political context and characteristics of the targeted institution, may also be used to understand and explain why institutional regulation may cause limited conflict within some industries and much more in others. Such differences between industries, combined with long-term changes such as changes in the labour market structure over time, for example an increase in the relative number of white-collar workers and a decrease in blue-collar workers, may thus contribute to institutional changes over time, where some norms and interpretations prevail over others.

References 

 177

References Bell, D. (1976). The coming of post-industrial society: A venture in social forecasting. Harmondsworth: Penguin. Crouch, C. (1993). Industrial relations and European state traditions. Oxford: Clarendon Press. Crouch, C. (1999). Social change in Western Europe. Oxford: Oxford University Press. Dahl, S.-Å., & Nesheim, T. (1998). Downsizing strategies and institutional environments. Scandinavian Journal of Management, 14(3), 239-257. Elster, J. (1992). Local justice: how institutions allocate scarce goods and necessary burdens. New York: Russell Sage Foundation. Falkum, E. (2008). Makt og opposisjon i norsk arbeidsliv. no. 128, Unipub, Oslo. Giddens, A. (1984). The constitution of society: Outline of the theory of structuration. Cambridge: Polity Press. Granovetter, M. (1985). Economic Action and Social Structure: The Problem of Embeddedness. American Journal of Sociology, 91(3), 481-510. Hagen, I. M. (1995). Hvem skal vi velge - en studie av to oppsigelsesprosesser [Who should we choose - a study of two dismissal processes]: Master thesis in sociology (cand.sociol). Institute for sociology and human geography. Universitetet i Oslo. Hall, P. A., & Soskice, D. W. (2001). Varieties of capitalism: The institutional foundations of comparative advantage. Oxford: Oxford University Press. Hult, C., & Edlund, J. (2008). Age and labour market commitment in Germany, Denmark, Norway and Sweden. Work, Employment & Society, 22(1), 109-128. doi: 10.1177/0950017007087419 Huzzard, T., & Nilsson, T. (2004). Dancing queen? Partnership, Co-determination and strategic unionism in Sweden. In R. Scott, T. Huzzard & D. Gregory (Eds.), Strategic unionism and partnership: boxing or dancing? (pp. 86-106). Basingstoke: Palgrave Macmillan. Immergut, E. M. (1998). The Theoretical Core of the New Institutionalism. Politics & Society, 26(1), 5-34. doi: 10.1177/0032329298026001002 Katzenstein, P. J. (1985). Small states in world markets: Industrial policy in Europe. Ithaca, N.Y.: Cornell University Press. Løken, E., Stokke, T. A., & Nergaard, K. (2013). Labour Relations in Norway Fafo-report 2013:09. Oslo: Fafo. Mahoney, J., & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining Institutional Change: Ambiguity, Agency and Power. Cambridge: Cambridge University Press. Mahoney, J., & Thelen, K. A. (2010). Explaining institutional change: Ambiguity, agency, and power. Cambridge: Cambridge University Press. March, J. G., & Olsen, J. P. (1984). The New Institutionalism: Organizational Factors in Political Life. The American Political Science Review, 78(3), 734-749. OECD. (2012a). Benefits and wages: Statistics. Retrieved 29.06.2012, from OECD OECD. (2012b). Synthetic Indicators of Employment Protection Retrieved 8. November, 2012, from http://www.oecd.org/els/employmentpoliciesanddata/oecdindicatorsofemploymentprotection.htm Oliver, C. (1991). Strategic Responses to Institutional Processes. The Academy of Management Review, 16(1), 145-179. Pfeffer, J., & Salancik, G. R. (2003). The external control of organizations: A resource dependence perspective. Stanford: Stanford Business Books. Pissarides, C. A. (2001). Employment protection. Labour Economics, 8, 131-159. Regini, M. (2000). The dilemmas of labour market regulation. In G. Esping-Andersen & M. Regini (Eds.), Why deregulate labour markets? (pp. 11-29). Oxford Oxford University Press.

178 

 References

Scott, W. R. (2008). Institutions and organizations. Thousand Oaks, Calif.: Sage Publications. Sels, L., & Van Hootegem, G. (2001). Seeking the Balance between Flexibility and Security: A Rising Issue in the Low Countries. Work, Employment & Society, 15(2), 327-352. doi: 10.1177/09500170122118986 Sippola, M. (2012). Local bargaining and codetermination: Finnish experience in comparative perspective. European Journal of Industrial Relations, 18(1), 53-69. doi: 10.1177/0959680111430564 Smith, M. R., Masi, A. C., van den Berg, A., & Smucker, J. (1995). External Flexibility in Sweden and Canada: A Three Industry Comparison. Work, Employment & Society, 9(4), 689-718. doi: 10.1177/095001709594004 Stinchcombe, A. L. (1968). Constructing social theories. New York: Harcourt, Brace & World. Stokke, T. A. (2008). The Anatomy of Two-tier Bargaining Models. European Journal of Industrial Relations, 14(1), 7-24. doi: 10.1177/0959680107086109 Streeck, W., & Thelen, K. A. (2005). Beyond continuity: Institutional change in advanced political economies. Oxford: Oxford University Press. Svalund, J. (2013). Adjusting labour through crisis: A three industry comparison. Economic and Industrial Democracy, first published online October 9, 2013 doi: 10.1177/0143831x13499619 Svalund, J., Casinowsky, G. B., Dølvik, J. E., Håkansson, K., Jarvensivu, A., Kervinen, H., Møberg, R. J., & Piirainen, T. (2013). Stress testing the Nordic models: Manufacturing labour adjustments during crisis. European Journal of Industrial Relations, 19(3), 183-200. doi: 10.1177/0959680113493838 Svalund, J., & Kervinen, H. (2013). Trade union power during labour adjustments – comparison of company-level cases. Transfer: European Review of Labour and Research, 19(4), 489-505. doi: 10.1177/1024258913501761 Wailes, N., Ramia, G., & Lansbury, R. D. (2003). Interests, Institutions and Industrial Relations. British Journal of Industrial Relations, 41(4), 617-637. doi: 10.1046/j.1467-8543.2003.00291.x Walton, R. E., & McKersie, R. B. (1965). A behavioral theory of labor negotiations: An analysis of a social interaction system. New York: McGraw-Hill. Walton, R. E., McKersie, R. B., & Cutcher-Gershenfeld, J. E. (1994). Strategic of labor negotiations: A theory ofchange in labor-management relations. Boston: Harvard Business School Press.

Heidi Nicolaisen, Sissel C. Trygstad

10 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market Dear customer, we are working on acquiring new customers who want to pay for our services legally (invoice). These customers give us the opportunity to conduct our business in accordance with Norwegian law. Each such customer is valuable to us. We have experience with private cleaning, offices, shops, staircases and other. If You know somebody looking for a cleaning company, we will be very grateful if You can recommend us. For each customer referral by you with whom we can cooperate, we offer you one FREE cleaning. Best regards, “Eva”.

Such was the text message that “Eva” sent to her customers after the Norwegian authorities introduced new “hard” regulations in the cleaning industry. The wage level in the collective agreement was extended to non-unionized workers by law, public approval was required for the cleaning companies and the employees had to carry ID cards. Eva referred to these changes as “a critical moment for business”. She feared even harder competition from companies at odds with the law. In this chapter we discuss the political efforts made to deal with precarious and marginal jobs in the cleaning industry in the period from 2010 to 2013. In Europe, scholars characterize the ongoing changes in working life as shifts towards dualization (Palier & Thelen, 2010; Thelen & Kume, 2006). Instead of total fragmentation and individualization of employment relations, dualization refers to a more polarized change process: Collective bargaining institutions erode in parts of the private service sector, while union density and membership in employer associations are still in place in manufacturing and the public sector. In Norway, a highly organized labour market is considered a cornerstone of working life and a key to maintain its ethos of equality, high-trust relations and inclusion (see, e.g., Gallie, 2003). In the cleaning industry many of the elements associated with the Norwegian model of labour relations1 are missing. After decades of growing concerns, the government and the social partners took several steps to improve the working conditions. Although the outcome was new and unusual, it derived from tripartite negotiations and cooperation. Rather than relying one-sidedly on the voluntary system of collective bargaining or campaigns with a soft and suggestive character,

1 In this chapter we apply the notion “labour relations” to refer to regulations (legislation, collective agreements, programs) and practices (routines for cooperation) governing the relationship between employers and employees (and sometimes government) at all levels. When we refer to the general analytical characteristics of the national system we apply “the Norwegian model of employment relations”. © 2015 Heidi Nicolaisen, Sissel C. Trygstad This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

180 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

hard regulations were introduced to combat illegal practices and to make the working conditions in the cleaning industry more decent. The term “decent” was launched by the United Nations ILO2 in 1999 to promote “opportunities for women and men to obtain decent and productive work, in conditions of freedom, equality, security and human dignity” (ILO, 1999, p. 3). Decent work refers to wages and working conditions. It also refers to employers who follow laws and regulations. The ILO’s “Decent Work Agenda” became a prioritized area for the Norwegian centre-left coalition government in 2009. The government’s effort to promote decent work was one of several factors that created a window of opportunity, a brief period in which opportunity for changes existed. This is, however, not a straightforward case of “soft regulations turning hard”. The regulations have traditionally consisted of a mix of soft and hard, which has been intertwined. The complex and co-dependent relationship between the Working Environment Act and collective agreements is one example (Stokke, 2004). Nevertheless, the emphasis has been on the soft side and the norm that the regulation of the labour market mainly is the domain of the voluntary system of collective bargaining. In this chapter we examine why the emphasis shifted towards harder regulations.

10.1 Dualization of Working Conditions and Labour Relations? In parts of the growing private sector services in Norway, the rates of organization among employers and employees are low, the coverage of collective agreements patchy and unions and legally proscribed institutions for employee voice and participation are often absent or declining (Longva, 2001; Nergaard & Stokke, 2007, Trygstad et al., 2011; Berge et al., 2013). Internationally, parts of private sector services, such as cleaning, are viewed as arenas where the hollowing out of the standard employment relationship have come furthest (Bosch, 2006, p. 44). In Europe, terms like “working poor” and “socially excluded” are used to characterize those situated in a permanently disadvantaged position in the labour market as well as in other spheres of social activity (Davidson & Naczyk, 2009). Hence, the development in Norway followed the same path as observed in several other European countries, where the rise of the private service sector is seen as one of several features that challenge and intensify the conflict between unions and employers. Sisson (2013) argues that the collective bargaining associated with trade union demands for a common rule for wages, hours of work and other terms and conditions of employment is strongly challenged, especially in the private service sector.

2 See the Report of the Director-General to the International Labour Conference meeting in its 87th Session.



Dualization of Working Conditions and Labour Relations? 

 181

When union density and membership in employer associations is concentrated in the public sector and manufacturing, and is in decline elsewhere, collective bargaining institutions will erode as employment shifts to the service sector (Thelen, 2012, p. 9). This process of dualization can be fuelled by an intensified cooperation between labor and management in core firms and industries which may leave other firms and workers “behind”, “outside” and in a dead end (Thelen & Kume, 2006; Palier & Thelen, 2010). Although immigration has accentuated the focus on bad working conditions in the cleaning industry, the problems have been debated in media and the Parliament since the mid-70s. The conditions in the industry have for decades caused concern due to the poor working standards. The question, however, is how can we explain that the social partner took action and implemented new hard regulations during the period from 2010 to 2012?

10.1.1 Methods and Data This chapter is based on two studies in the cleaning industry in 2011 and 2012 (Trygstad et al., 2011; Trygstad et al., 2012). In both studies semi-structured interviews were conducted with the social partners at the company level, trade union officials, safety delegates and employers in cleaning companies in Oslo, Bergen and Trondheim and representatives from the police, the tax authorities and the Labour Inspection Authority. A total of 39 interviews were conducted. Moreover, the studies where based on thorough analyses of Statistics Norway’s structural statistics, where we examined company structures as well as employee characteristics. The 2011 study included structured interviews with 40 purchasers of cleaning services in the private and public sectors, while the 2012 study was based on a telephone survey of employers, managing directors or similar in cleaning companies with five or more employees. Key issues for the survey were the companies’ experiences related to the extension of the collective agreement, competition from actors at odds with the law and the instruments established to provide more orderly conditions. Altogether 205 employers/managers responded to the survey, which hereafter we will refer to as “the cleaning-industry company survey” (for further details, see Trygstad et al., 2012). In the following section we outline the analytical perspectives on institutional change before we address the foundations of the traditional system of Norwegian industrial relations and how the cleaning industry fits into this general picture. In section four we present the institutional measures introduced to combat bad jobs in cleaning and discuss their capacity to prevent further deterioration of working conditions in the sector. In section five we discuss how we can understand the regulative change and why a “window of opportunity” opened in this period. We also discuss whether hard regulations may be seen as a departure from the voluntary system associated with the Norwegian model of labour relations. In this period

182 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

industry actors as well as the Ministry of Labour united their political power, resources and will to produce substantial change. Finally, we discuss the findings and arrive at a conclusion in section six.

10.2 Analytical Perspectives on the Prospects of Change The effort to improve the working conditions in the cleaning industry can be seen as a dramatic institutional change because of the methods used, for example, the extension – by law – of the wage level in the collective agreement to non-unionized workers, in contrast to the traditional emphasis on voluntary collective bargaining. According to Sisson (2013) legislative initiatives can be tailored to ensure changes. In this case such initiatives were implemented to ensure what Gallie (2007 pp. 17-19) describes as an inclusive employment regime. He argues that inclusive employment regimes are designed to extend employment and common employment rights as widely as possible, whereas dualist regimes guarantee strong rights to a skilled, core workforce at the expense of the poor conditions and low security of the periphery. A  third system, market employment regimes, emphasizes minimal employment regulation and assumes that market adjustments will ensure relatively high employment levels. In Norway, a package of hard regulation was introduced in the cleaning sector. We define hard regulations as statutory regulations that are enforced and controlled by the Labour Inspection Authority or tax authority. By introducing hard regulations, the development in Norway takes another path than what is registered in several other European countries, where hard regulations have been challenged by soft “framework agreements”. The purpose of such agreements is to establish broad principles and incite negotiations or “joint declarations”. Sisson (2013) describes the last one as advisory and its implementation is not addressed. To simplify we use soft and hard regulations as a starting point in the table below, where we also summarizes possible outcomes on working conditions due to different regulation approaches: Table 10.1: Quality of working conditions (WC) and different types of regulation Quality of working conditions Poor WC

Type of regulation

Soft Poor WC because of lack (voluntary and high level of employment relations of discretion) and regulations (at the company level) Hard (statutory and low level of discretion)

Decent WC Decent WC because of high quality employment relations and regulations (at the company level)

Poor WC because of lack Decent WC because of of enforcement/control rigorous enforcement/ control



Analytical Perspectives on the Prospects of Change 

 183

The organized part of the Norwegian labour market is characterized by high quality working conditions regulated primarily by soft measures – a voluntary system of regulations where key features are high coverage of collective agreements, high trade union density and highly organized employers. At the company level shop stewards, safety delegates and managers are involved in tight formal and informal cooperation. The situation is otherwise in the cleaning industry, characterized by weak employment relations and in large parts poor working conditions. The hard regulations introduced are characterized by a low level of discretion and very limited flexibility. The need for hard regulations may be seen as the two parties being unable to solve the challenges by themselves, due to a lack of power resources. Findings from studies in other industries indicate that the confluence of different measures has helped establish an institutional framework that can effectively counteract social dumping and dualization (Eldring et al., 2011). Ignoring a scheme or a provision is more risky if the changes that are introduced pull in the same direction and if their rationale is in accordance with established norms and values (Oliver, 1991, pp. 162-163). Oliver (1991) argues that companies are more likely to ignore a new scheme or structure if the organization can argue that the “likelihood of “getting caught” is very low or that its success is not dependent on government approval and support” (Oliver 1991, p. 156). Moreover, time is an element in institutional change processes (Tolbert & Zucker, 1996). At an initial stage of change control will be crucial to enforce new regulations because norms and practices have not yet developed. The power and interests of the actors involved is another key issue. Several scholars in the field of institutional theory stress that the establishment of viable institutions or institutional structures is a demanding task (Selznick, 1957) that require a form of “collective action” where actors invest material and cultural resources in the new institution (Pierson, 2000, p. 258). The willingness to invest in such action presupposes some level of shared interests. The required collective effort can be impeded if the actors have opposing interests. Institutional transformation is thus dependent on having a critical mass of supporters of the new structure at the initial stage. The success of the measures cannot be taken for granted; the process may grind to a halt or change direction along the way because of failing support and legitimacy. The ability of losers or opponents of an institution to halt, slow down, undermine, openly counteract or in other ways change the process will depend, among other things, on their alternative strategies and power resources. Mahoney and Thelen (2010, p. 23) describe various types of actors: insurrectionaries, symbionts and/or opportunists. In this context, it is crucial to identify the preferences and strategies of the actors. Mahoney and Thelen (2010) distinguish between changes driven by strategic actors, changes in the environment and changes following from unintended consequences. Moreover, a potential for change may be latent in how institutions/ rules are ambivalent, unclear or contested. There are, moreover, examples of situations completely devoid of rules or where new rules replace existing ones (“displacement”) or where rules “drift” because they are not adjusted to development

184 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

in the organizations. The outcome of the various processes of institutional change will depend on different actors’ willingness, interests and power to follow the changes introduced, and vice versa – to avoid them. In our context the change capacity of the hard regulations may depend on the permanent provision of statutory regulations, control and enforcement. An alternative route to decent working conditions would be that a regime of voluntary soft regulations emerges and over time reduces the need for hard regulations.

10.3 The Foundation of the Norwegian Industrial Relations Model The Norwegian model of labour relations consists of a combination of regulations embedded in collective agreements (voluntary regulations) and statutory regulations. The model combines “hard” and “soft” dimensions (Marginson & Sisson, 2004, p. 87). The basic protection of employees is ensured through a comprehensive Working Environment Act (WEA) with provisions for health, working environment and safety, working hours, hiring and dismissal, etc. The Act signals a high level of ambition as regards the physical and organizational working environment and stipulates that the employer in cooperation with the company’s safety organization is responsible for this. A key remit of the monitoring authorities (the Labour Inspection Authority) is to assist the companies in facilitating appropriate routines for improving the working environment and to monitor compliance. Several of the provisions in the Norwegian WEA allow flexibility in implementation through collective agreements. Collective labour relations and collective agreements fulfil a key function in the regulation of wages and working conditions (Nergaard, 2014). First, collective agreements play a major role in regulating wages, since Norway has no statutory minimum wage.3 In cases where issues are regulated by law and collective agreements, for example overtime pay and holidays, the latter normally provide more favourable conditions for workers. Second, co-determination in the workplace primarily occurs through the elected trade union officials in the workplace and the rules for provision and consultation of trade union officials are laid down in the collective agreements. The Norwegian representation system is based on so-called single-channel representation, meaning that the representation at the workplace level is based on representatives of the trade union organizations. In practice, this type of trade union official will also be involved in matters pertaining to the working environment and working conditions, and in some cases the safety delegates will be active shop stewards or “dedicated members” of the local trade union (Falkum et al., 2009). Third, effective implementation of the employees’ rights will in many cases rely

3 In the EU, 21 of the member state have a statutory minimum wage. This form of statutory minimum wage is not found in the Nordic countries (Alsos & Eldring, 2012, p. 8).



The Foundation of the Norwegian Industrial Relations Model  

 185

on the presence of a trade union and cooperation between the local union and the management (Dølvik et al., 2014; Hagen & Trygstad, 2009). This applies particularly to employees who have a weak bargaining position or are unaware of their rights. Thus, the two sets of regulatory institutions (voluntary regulation and statutory legislation) may be seen as co-dependent. It could be argued that decent work is embedded in the Norwegian model of labour relations and constitute a “logic of appropriateness” that affects all actors involved (Scott, 1995, p. 39).

10.3.1 The Cleaning Industry – on the Margins of the Norwegian Model of Labour Relations In the context of the generally well regulated, decent and egalitarian Norwegian working life there has over the last four decades been a growing concern for the cleaning industry. The challenges are partly associated with the structure of the industry and partly with a changed workforce, weak collective labour relations and increased competition.

10.3.2 The Structure In Norway, as well as in other countries, cleaning it is characterized by many small companies and rapid turnover in terms of companies as well as activities (O’Sullivan & Turner, forthcoming, Trygstad et  al., 2012). From 2000 to 2010, the number of registered companies grew by 60 per cent (Trygstad et al., 2012, p. 33). More than half of them did not survive for more than one year (ibid., p. 40). One reason for the rapid turnover is the low start-up costs. Only minimal investment is required to establish a company. The low start-up costs have positive as well as negative aspects. On the positive side, they help create alternative career pathways. On the negative side, the minimal obstacles to establishment make bankruptcy less risky, for example because of non-payment of taxes, duties and/or wages. A reputation as a dishonest operator can easily be bypassed by establishing a new company under another name. These features challenge institutionalization, transparency and control.

10.3.3 The Workforce Transparency and assertion of interests in the industry are also hindered by the low skill requirements for employers and employees. Moreover, wage-related costs amount to 85 per cent of the costs in cleaning (NHO Service, 2009) and thus affect the recruitment strategy of employers. The cleaning industry was originally dominated by housewives working part-time, most often during afternoons and evenings. During

186 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

the last decades, cleaning has been transformed and institutionalized more in line with other parts of the labour market when it comes to standardized working hours. Cleaning is more often characterized by daytime work, and the proportion of part time workers with short hours are decreasing (Trygstad et al., 2012). A further sign of the formalization process is the possibility to take a craft certificate. At the same time, the industry is still characterized by manual work and low skill requirements, which make the cleaning profession a gateway to the labour market for immigrants with limited language skills. The proportion of workers with an immigrant background has increased in the industry through the 2000s, and workers from the new EU countries make up one-quarter of the registered employees whereas almost 40 per cent have an Asian or African background. This group often has very limited knowledge about the formal regulations of Norwegian working life. Nor do they have many career alternatives. Hence, these employees are vulnerable to employers who violate laws and regulations or who wish to minimize the costs of wages and social benefits.

10.3.4 The Labour Relations A third circumstance that has driven the cleaning industry to the margins of the Norwegian model of labour relations is the absence of strong collective labour organizations. Few employers are organized, and the unionization rate among employees is low. Thirty per cent of all employees work in companies with no or very weak labour organizations, and safety delegates are quite rare (Berge et  al., 2013). Even in companies with a collective agreement, recruiting trade union officials may be a problem. Often the companies have no tradition of social partnership or co-determination, and the management and the trade union officials tend to meet only when issues of a negative nature have to be discussed. Moreover, conditions for performing the trade union tasks are poor. The representatives seldom have sufficient time to attend to their elected office, and they do not meet their members on a daily basis since these are dispersed in the client companies. Hence, the capacity of trade union officials and safety delegates to oppose management and act as partners in the development of good-quality workplaces is largely absent.

10.3.5 The Competition The increasing competition for assignments is the fourth challenge characterizing this industry. The cleaning companies receive many of their assignments through competitive tendering, and through these the clients shape the frameworks of the industry. The client designs the tender, selects the supplier and signs the contract. The client will also monitor compliance with the contractual requirements. However,



The Foundation of the Norwegian Industrial Relations Model  

 187

the competitive tendering has consequences, and some of the actors in the industry are concerned about the increasing work intensity: Performance is going up and up. In the 1990s we talked about average performance. Then, 250 square metres per hour was seen as the maximum limit. Today, nobody bats an eyelid if we exceed 500 or approach 1000. For many years, there was a very positive development... it could be defended because of more efficient methods. Competition was healthy. But then we reached a saturation point. (Employer, quoted in Trygstad et al., 2011, p. 109)

This price pressure is partly ascribed to highly price-sensitive clients and partly to suppliers who provide extremely cheap services by cutting corners and/or violating laws and regulations.

10.3.6 Different Degrees of Decency

The companies in the cleaning industry can be simply divided into three groups, where the decent ones are challenged by the companies in the two other categories when it comes to competing over contracts and market shares. The decent: Comply with legislation, rules and agreements.

↔

Intermediate category: Operate partly within and partly outside of legislation, rules and agreements.

↔

The not decent: Operate systematically outside of legislation and rules.

Figure 10.1: Model of the degree of decent working conditions. Source: Trygstad et al., 2011.

The arrows between the boxes illustrate that companies may be pushed in a different direction as a result of changed regulations. Companies in the intermediate category will typically not have a collective agreement, and they will use sub-contractors and self-employed workers and other forms of precarious work extensively to cut costs. They will also increase work intensity by pricing assignments at a low level. Among the not decent operators we will find workers who are grossly exploited and work illegally and companies that systematically violate regulations on taxation and VAT. Furthermore, these companies have established complicated company structures to conceal illegal activities, including money laundering (Trygstad et  al., 2011). A critical issue will be whether the measures implemented will be able to push parts of the intermediate category in a more decent direction and to make the situation so untenable for the not decent ones that the number of such companies decreases. The intermediate category will thus comprise a type of actor who more or less deliberately and strategically violates the generally accepted norms of appropriateness that

188 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

characterize the Norwegian working life and thus undermines institutions intended to ensure decent conditions in the cleaning industry. According to informants, the intermediate category accounted for most companies prior to the introduction of hard regulations in 2011 and 2012 (Trygstad et al., 2011, 2012). One employer commented on the situation: Things develop so quickly. The decent companies are forced into a grey zone. If paying below tariff wages is legal and the rest of the industry moves in that direction, we have to do the same in order to survive. I’m afraid that we are moving the boundaries of what is deemed to be acceptable. (Employer, the cleaning industry, quoted in Trygstad et al., 2011, p. 175)

Seen from the perspective of the cleaning industry, dualization can thus be seen as a process transforming both working conditions and industrial relations institutions. This process differs from the much debated concept of labour market deregulation. According to Thelen (2012, p. 9) deregulation proceeds through a direct assault on traditional institutions, while dualization often proceeds through a process of institutional drift. The way the hard regulation works will affect how the institutionalization process moves forward.

10.4 Measures to Prevent Dualization The new institutional measures introduced to combat bad jobs in cleaning held a strong focus on customer awareness. Those who procure cleaning services in the private-company market are made more responsible with regard to ensuring that the suppliers operate in line with the regulations in laws and agreements. In the public sector clients have had such a responsibility for years. Pursuant to the Public Procurement Act a set of regulations for wage levels and working conditions was adopted in public contracts, first for government companies in 2005 and from March 2008 in public contracts (the entire public sector). The regulations state: In their contracts, the commissioning agency shall include the requirement that employees of suppliers and any sub-contractors that are directly involved in fulfilling the contract shall receive wages and have working conditions that are not less favourable than what follows from nationwide collective agreements or what is otherwise normal for the location and profession in question.4

Nevertheless, there have been several reports of violations of the ILO convention (Trygstad et  al., 2011). Problems have particularly been associated with controlling employees’ work conditions when the cleaning companies apply sub-contractors.

4 Regulation No. 112/2008 on pay and working conditions in public contracts.



Measures to Prevent Dualization  

 189

The new measures in the cleaning industry have highlighted the responsibility of the clients. This can be regarded as an admission of the inability of the social partners and the authorities to increase compliance in the industry unless the clients shoulder their share of the responsibility. In the following sections we elaborate on further measures introduced to prevent a “race to the bottom” in this industry.

10.4.1 Extension of Collective Agreements Norway does not have a general minimum wage, but an equivalent was established in the cleaning industry as the wage level in the collective agreement was extended by law to non-unionized workers on 1 September 2011. The regulations pertain to private companies that sell cleaning services and to employees who perform cleaning work. The wage rates are regulated in line with the revisions of the collective agreement. Self-employed persons are not covered by this extension. This may have substantial consequences as one in four cleaning companies staff assignments with self-employed workers (Trygstad et al., 2012).

10.4.2 Approvals and ID Cards In September 2012, a system for public approval of cleaning companies was introduced.5 This approval system includes ID cards for the employees. Moreover, companies must be approved by the Labour Inspection Authority to obtain legal status, and purchasing cleaning services from non-approved companies is illegal.6 Approved companies shall be listed in a registry maintained by the Labour Inspection Authority and are also obliged to issue ID cards for their workers. To obtain approval, the following conditions must be met (cf. Section 5 [1])7: –– Provision of approved company health services. –– Provision of a safety organization.8 –– Written labour contracts. –– Compliance with the wage level in the collective agreements. –– A scheme that ensures financial compensation to workers in case of occupational injury.

5 FOR 2012-05-08-408: Authorisation scheme for cleaning companies (Forskrift om offentlig godkjenning av renholdsvirksomheter og om kjøp av renholdstjenester). 6 i.e. private individuals are not required to check whether the enterprise is approved if the cleaning services are purchased for private use. 7 These requirements do not apply to self-employed workers. 8 In enterprises with less than 10 employees, the partners can agree to waive the requirement for a safety delegate.

190 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

10.4.3 A Mix of Soft and Hard Regulations According to Table 1, the new measures introduced could be characterized as hard regulations based on rather extensive control from governmental institutions and harsh sanctions for those who violate the new provisions. To consider the development in the industry as a drift away from the traditional “voluntary system” associated with the Norwegian model of labour relations would, however, be a simplification. As we will elaborate below, the social partners were deeply involved in the establishment of a tripartite industry programme for the cleaning industry. The process preceding the introduction of the measures can be described as a typical for Norway; it was “voluntary” and “soft” and characterized by cooperative partners and a great deal of consensus, but the chosen tools were hard. This brings us to examine further the second research question: To what extent do these measures harmonize with the traditional “voluntary system” of industrial relations associated with the Norwegian model, and what enabled the introduction of hard regulations?

10.5 The Introduction of Hard Regulations – A Window of Opportunity The question raised is why the social partners took action and implemented hard and new measures during the period from 2010 to 2012. The period represented a window of opportunity9 where industry actors as well as the Ministry of Labour united their political power, resources and will to produce substantial change. Why did this happen at this time? As already mentioned, concern for the cleaning industry has been on the political agenda more or less since the 1970s. In 1978 the social partners’ agenda was to improve the working environment, create safe and stable workplaces and establish satisfactory wage levels. However, nothing happened (Trygstad et al., 2011). In 1993, a voluntary approval scheme was established for companies in the industry. The intention was that “Clean Development” should induce the companies to comply with the provisions of the Working Environment Act and the Taxation Act. The introduction of craft certificates in 1995 was another example of institutional efforts aiming at improving wages and working conditions (Skilbrei, 2009). The introduction of “Clean Development” and craft certificates can be regarded as an attempt to achieve decent working conditions through “soft regulation” (Sisson & Marginson, 2001), but it had few results. The problem of dishonest companies did not abate. There may be several explanations for the window of opportunity that opened up for the introduction of hard regulations during 2010-2012. Key actors had mutual interest in change, something which is a critical factor in a process of

9 Concept used to describe a brief time period in which an opportunity exists



The Introduction of Hard Regulations – A Window of Opportunity  

 191

institutionalization (Mahoney & Thelen, 2010; Selznick, 1997). At the same time the situation in the cleaning industry reached the public through media, and the lack of control with the business speeded up the recognition that harsh measures, control and sanctions for those violating the rules was needed. Political willingness and a government which is indebted to the trade union movement: ILO’s “Decent Work Agenda” was the basis for the centre-left government’s 2009 programme to reinforce and coordinate efforts for workers’ rights internationally and at home.10 The centre-left coalition government’s concern with the development in the industry spurred the establishment of a tripartite industry programme for the cleaning industry in the context of the wage bargaining in 2010. The intention was to create a long-term effort to improve working conditions, co-determination, working environment and safety (Melding til Stortinget nr. 1 2010-2011, p. 19). The government’s efforts should also be seen in relation to its attempts to combat social dumping in industries characterized by immigrant labour and to avoid growing tendencies of dualization in the labour market at large. At the same time, this programme represents something new. It constitutes a broader and more thorough approach compared to previous programmes that mainly targeted immigrants in short-term and long-term contracts in industries that otherwise were well organized and regulated. Escalation of problems and public condemnation: From 2007 and onwards, the media intensified its searchlight on the cleaning industry. The number of media search hits for the phrase “social dumping AND cleaning” multiplied (Trygstad et al., 2011, p. 33). A main reason was the increasing labour immigration from Eastern Europe. The proportion of Eastern Europeans in the cleaning industry increased from seven to 22 per cent between the EU enlargement in 2004 and 2007 (Trygstad et al., 2011, p. 82; Trygstad et al., 2013, p. 61). Recurrent headlines were “slave contracts”, “human trafficking”, “illegal work”, “tax evasion” and “inhuman wage levels and working conditions”. Many of the reports concerned cleaning companies engaged by public agencies. The demands for better control and a clean-up of the industry grew and were backed up by arguments such as “competition on equal terms”, “decent pay and working conditions” and a greater degree of “decency” – key norms and values in Norwegian working life. These are also value-laden concepts that can hardly be opposed. This attention and condemnation in the public stimulated the rather dramatic institutional changes. Like-minded and willing partners: Over time, the key actors in the cleaning industry have developed a shared understanding of the need for change. Our findings indicated a very positive personal chemistry among central actors. This is one of the factors that made changes possible. Although their motives differed somewhat, the actors also shared an interest in the outcome of change. The trade union

10 http://www.regjeringen.no/nb/dep/ad/pressesenter/pressemeldinger/2008/strategi-for-et-anstendig-arbeidsliv.html?id=525795

192 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

movement wanted to improve wage levels and working conditions and counteract social dumping. The employers were keen to establish competition on more equal terms and to restrict the number of companies operating at the edge of the law. This situation resulted in a collaboration that survived a number of challenges, including explicit disagreement over certain measures, nationwide strikes in 2010 and 2012 and internal disagreement on the employers’ side in regard to how far the service-industry union in the Confederation of Norwegian Companies (NHO) should go in accepting controversial measures. New instruments had become available: Increasing labour immigration from the new EU countries had already changed working conditions in other industries where the social partners had held a strong position (such as construction work). The trade union movement had pressed for the introduction of new measures. The extension of the wage level in collective agreements was used for the first time in 2005 and was later applied to several industries with a large proportion of migrant labour. Such schemes were previously only applied when it was proven that foreign workers had less favourable conditions than Norwegian workers, but they nevertheless established a minimum wage level in an entire industry. Hence, the partners in the cleaning industry had available a set of institutional instruments that had been partly or fully tested in other industries. The construction industry had issued ID cards to employees for several years, and a registration scheme for man-power suppliers had been introduced to ensure better control and transparency in the companies. It can be argued that years of (voluntary) soft regulation and self-regulation had failed to counteract a race to the bottom in the cleaning industry. The combination of public attention towards the industry (partly because of increased labour migration) political determination and unified social partners created a pressure for change that was hard to resist. This paved the way for “hard” measures in an industry that was on the verge of collapsing. To curb companies operating at odds with the law, and to prevent drift towards further dualization in working conditions, the parties agreed to unconventional measures. The introduction of new institutional measures, however, brings us to the third research question: How can we understand the new measures’ capacity to change working conditions?

10.5.1 The Change Capacity of Hard Regulations – The First Experiences The ability of these hard regulations to develop into viable institutions depends on sufficient support from the actors. The social partners and the government must be motivated to continue investing in these hard regulations. Their motivation for such conduct probably depends on the initial experiences. A critical factor is whether the hard regulations actually are able to push the intermediate category of companies, as shown in Figure 1, in a more decent direction. These companies operate partly within and partly outside of legislation, rules and agreements. Another question is related to



The Introduction of Hard Regulations – A Window of Opportunity  

 193

the more general working conditions in the decent part of the industry where work is often demanding although it is conducted in accordance with the law. As the changes examined were recently introduced, data and analyses are few and mostly cover the actors’ expectations with regard to outcomes. We conducted interviews prior to the implementation of the measures in 2011 and found great expectations in the companies and in the industry in general. The actors had high hopes for the efficiency of the combined measures. Our informants also stressed the need for control and sanctions to ensure compliance. One key reason for the call for control is the increased competition from dishonest suppliers over recent years. In the company survey from 2012, 55 per cent responded that they had often lost tenders to such actors. In the same survey, company managers were asked to rank the measures they believed would be effective. The first two questions addressed expectations for the approval scheme and the ID cards, and the results are shown in Figure 2.

Approval scheme

3 4

ID cards

4 2

0% Not sure

13

26

8

10%

Not at all

54

24

20%

30%

To a small extent

61

40%

50%

60%

To some extent

70%

80%

90%

100%

To a large extent

Figure 10.2: To what extent do you believe that the approval scheme for cleaning companies and the requirement to hold ID cards will improve the conditions in the industry? Percentages. N=205.

Eighty per cent of the managers expected the approval scheme to improve the conditions to a great extent or to some extent. Altogether 85 per cent of the managers believed in the introduction of ID cards. Only very few believed that these measures would have no effect and a few were uncertain. There are no significant differences between companies with and without a collective agreement. A key point is that the approval scheme and the ID cards will have a combined effect, since the ID cards are linked to the approval scheme. The company managers were also asked to assess to which extent other measures would combat dishonest practices. Options included continued extension of the

194 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

collective wage agreement to non-unionized workers, more monitoring by clients, more monitoring by the tax authorities and more frequent inspections by the Labour Inspection Authority. The company managers believed that all these measures would be important (between 80 and 90 per cent). However, more monitoring by the tax authorities was ranked at the top of the list.

10.5.2 Something Has Already Worked The extension of the wage level in the collective agreement to non-unionized workers has been in effect since 1 September 2011, and in the spring of 2013 it was decided to prolong it for another two years. In our survey of managers in cleaning companies we also asked about their preliminary experience with the extension of the collective agreement. Fully 92 per cent responded that they had good (57 per cent) or fairly good (35 per cent) knowledge about this provision. Managers in large companies (more than 50 employees) are more knowledgeable than others in this respect. The managers were also asked whether the extension of the collective agreement to non-unionized workers has had an impact on the wage level in their company. In four out of 10 companies, the employees received higher pay as a result of the decision. Altogether 29 per cent of the managers responded that the majority of the employees had received a pay rise, and 11 per cent responded that some of the employees now receive more pay (not shown). It is interesting to examine whether there are any differences between companies with and without a collective agreement, as shown by Figure 3.

Without collective agreement

36

Collective agreement

21

0%

10%

11

47

12

20%

The majority of the employees received higher pay Have not had any impact

30%

6

64

40%

50%

60%

3

70%

80%

90% 100%

Some of the employees recieved higher pay Not sure

Figure 10.3: Has extension of the collective agreements wage level to non-unionized workers had an impact on the wage level in your company? Percentage. N = 205.



Discussion and Concluding Remarks  

 195

At the outset, one would not assume that the wage level should increase to any great extent in those companies that had a collective agreement, since the extension pertained to tariff wages with 0-2 years of seniority. In total, more than 30 per cent of the companies with an existing collective agreement have needed to raise the pay for the majority of the employees (21 per cent) or for some of them (12 per cent) as a result of the regulations. Unsurprisingly, in companies with no collective agreement the proportion is higher: 36 per cent of these responded that the majority of the employees have received a pay rise, and 11 per cent reported that some of their employees now receive better pay. This survey gives us no information on the size of these pay rises. Our data show, however, that a large proportion of the cleaning companies have been forced to adjust their pay levels since the extension of the collective agreements was enacted. Another potential effect regards client behaviour. A little more than half of our respondents said that the clients to a greater extent requested documentation on wage levels and working conditions in 2012, compared to the situation in 2010 (Trygstad et al., 2012, p. 108). The role of the clients is important to combat poor working conditions and dishonest practices. Client attitudes also play a role in improving working conditions in the legitimate part of the industry (the decent category), although the new regulations were primarily targeted at combating practices at odds with the law. Thus, our analyses indicate that the introduction of hard regulations has had effects on wage level and the customers’ awareness. It is reasonable to believe that these regulations have made it more difficult to operate partly within and partly outside of legislation, rules and agreements. This is important because it may reduce the number of companies operating at odds with the law and also make it much more complicated for those who systematically operate outside of legislation and rules. Moreover, we have argued that a process of institutionalization requires that new measures and regulations be supported. This support is essential, particularly at an early stage. If the companies and the employees in the decent part of industry experience that the regulations actually make a difference, this may extend the support and hence improve working conditions further.

10.6 Discussion and Concluding Remarks The changes introduced between 2010 and 2012 mark a divide in the approach to combating social dumping and the dualization of working conditions in the Norwegian labour market. We have defined the new measures as hard regulations, differing from the soft and voluntary regulations that normally characterize the Norwegian model of labour relations. Hard regulations enforce decent pay and working conditions by statutory regulation, limited discretion and intensified controls conducted by tax authorities, the labour inspectorate and other state authorities. Soft regulations, on the other

196 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

hand, are based on a voluntary system where decent working conditions are achieved through collective bargaining and cooperation among the social partners at the company and industry levels. In Norway, as well as in the other Nordic countries, soft regulations function well in the organized labour market. But in cleaning, the collective actors are relatively weak. This makes soft regulations insufficient to bring forward more decent working conditions. Soft regulations, which Marginson and Sisson (2004, p. 87) exemplified with “framework agreements”, have a high level of discretion and flexibility, but most sanctions have brought forward necessary changes. However, in the new system soft and hard regulations are tangled in several ways. Hence, it is problematic to characterize the change process as a shift from one system to another or as institutional conversion (Mahoney & Thelen, 2010). Several of the new hard regulations derived from tripartite negotiations and cooperation – institutional elements rooted in the Nordic model of labour relations. Rather, the shift in methods to achieve more decent working conditions is a case of institutional displacement, where the partners involved grasped the opportunity to introduce considerable change in methods. When it comes to the capacity of hard regulations to hinder poor working conditions, we have to include actors, power and interests into the discussion: trade unions, employer organizations, the government, the customers and the companies. Mahoney and Thelen (2010), more than conventional institutional theory, emphasize the role of strategic actors. They argue that institutional continuity or change cannot be understood as results of actors who passively reproduce institutional rules. Rather, institutions should be seen as emanating from interest mobilization, power struggles and compromises between strategic actors with partly opposing interests, whose actions continuously undermine, oppose, reproduce or convert the institution. In our case, the viability of the hard regulations depends on the interests and the choices made by the actors involved. The trade unions and employer organizations are core actors in the social partnership in the cleaning industry. They agreed on hard regulations on slightly different, but related, premises. While the trade unions were motivated primarily by the chance to improve working conditions, the employers were mostly concerned about unfair competition from the indecent companies. The strength and viability of this compromise depend on continuous support from both sides, but there are several challenges. One example is conflicting interests on the employer side. The proposal for extension of parts of the collective agreement to non-unionized workers was met with fierce resistance in the central organization of the Confederation of Norwegian Companies (NHO). On the other hand, the union of service industry employers, of which most of the organized employers in the cleaning industry are members, welcomed the proposal (NHO Service). To the trade unions, the extension of collective agreements is a double-edged sword. On the one hand, this is crucial to combat social dumping. On the other hand, it robs the union of a key argument to raise the unionization rate. The argument that membership in a trade union will give



Discussion and Concluding Remarks  

 197

higher pay is nullified when the employees receive the generally applied wage rates anyway. The Norwegian Confederation of Trade Unions (LO) has been sceptical of the request for an extension of the wage rates in the collective agreement. The union dominant in the industry, Norsk arbeidsmandsforbund, did however consider this as a useful means to combat social dumping. At the same time, both sides – employers’ and employees’ organizations – fear that the extension of collective agreements shall infect the labour market as such. A shift (back??) from hard to soft regulations in the cleaning industry will probably rely on the two parties’ capacity to organize labour and employers in the years to come. Another core actor is the government. Government intervention is important to regulate those parts of the labour market where there is less partnership, since the partners themselves will be unable to compel the actors at the company level to abide by the new rules (Traxler, 2003; Howell & Givan, 2011, p. 251). Continuity in a regime of hard regulations relies on rigorous control by the Labour Inspection Authority and tax authorities, which in turn rests upon the political willingness to invest resources in such control. In this case the government played a crucial role in establishing a regime of hard regulations with the rationale of combating social dumping and further dualization of the Norwegian labour market. Across countries it is observed that the government’s role in regulating working life varies in accordance with shifting political constellations in power (see, e.g., Traxler et al., 2001, pp. 155-156). After the election in autumn of 2013 the “red-green” government was replaced by a “blue-blue” coalition government with the Conservative party and Progress party. So far the new Minster of Labour has flagged a general willingness to continue the line established by his predecessors with regard to regulating and control of the service industries. The interests and strategies of the cleaning companies are also of great importance to understand the change capacity and viability of the hard regulations. The majority of Norwegian companies comply with the institutionalized norms and rules of working life. This cannot be explained by “habit”, but rather as a deliberate and strategic choice to follow institutional guidelines motivated by a desire to ensure availability of resources, predictability and/or social support (Oliver, 1991, p. 153). For companies in the intermediate category, avoiding new rules by buffering themselves from institutional pressures or escaping from institutional rules or expectations may represent a solution (ibid., p. 154). One way to avoid the increased wage costs that follow from the extension of the collective agreement is to reorganize the company into a “service contractor” using self-employed workers instead of regular employees. In doing so, the extension of the collective agreement is circumvented. An increasing number of self-employed workers in this industry may indicate that the change capacity of the hard regulations is low because company actors find ways to “avoid” (Oliver, 1991) or “convert” (Mahoney & Thelen, 2010) the rules by exploiting loopholes in the regulations. The formal rules remain intact, but they are interpreted in an opportunistic manner that undermines them. Another company strategy may be to defy the new rules by ignoring them. Companies that fail to apply for approval

198 

 Preventing Dualization the Hard Way – Regulating the Norwegian Labour Market

or issue ID cards to their employees, or who deliberately continue to market their services without approval, exemplify such defiance of the rules. The rationale for such conduct may be that the likelihood of getting caught is low, or that the company is independent of government approval and support (Oliver, 1991, p. 156). However, this is conditional on there being no reactions from their environment. Moreover, the decent companies would also have to support the new regulations in order for them to work efficiently. There is, for example, a widespread expectation that the generally applied hourly rates will result in intensification of work. To combat such outcomes at the company level, cooperation between management, safety delegates and trade union officials is crucial. This, however, requires that the trade union officials and safety delegates are allocated sufficient time to attend to their offices and remit and that the employers actively seek consultation with the employee representatives. The customer is the last actor affecting the change capacity of the new hard regulations. It is not unreasonable to assume that some customers still will want to purchase services at the lowest possible price, even illegally if need be. Thus, the change capacity of the hard regulations relies on their ability to change the norms and behaviours of the customers. Customers would have to balance concerns for the quality and working conditions against the price more evenly than have been seen until now (Berge et  al., 2013). “Eva”, whom we quoted in the introduction, is an employer who encourages her customers to comply with the new industry rules and Norwegian law. While ideals of decent working conditions seem to enjoy general support on an ideational level, it remains to be seen whether their normative force is strong enough to combat unilateral low-price competition. The introduction of hard regulations is part of an effort to avoid further dualization of the Norwegian labour market. But, the working conditions and industrial relations in the cleaning industry is not an example of a “good case turning bad”. The industry has for several decades been marked by precarious work and low coverage of collective agreements. One could argue that the labour migration from Eastern Europe only accentuated old and well known problems in the industry. This may be seen as a case where the institutionalized “norms of appropriateness” – deeply rooted in Norwegian labour market – caught up with the social partners and the government and created a window of opportunity. Although the hard regulations do not harmonize with the preferred methods in Norwegian labour relations, they may succeed in avoiding the industry-specific race to the bottom seen across Europe and foster a more inclusive employment regime. So far, the measures introduced have resulted in a wage increase. However, the changes have not as of yet hugely transformed working conditions in the industry. Rather, we are witnessing an example of quite large and partly transformative institutional change resulting in – so far – small improvements.

References 

 199

References Alsos, K., & Eldring, L. (2012). European minimum wage: A Nordic outlook. Oslo: Fafo-report 2012, p. 16. Berge, Ø. M., Nergaard, K., & Trygstad, S. C. (2013). Anstendig arbeid. Fafo-rapport 2013:03. Oslo: Fafo. Bosch, G. (2006) Working time and the standard employment relationship. In J. Y. Boulin, M. Lallement, J.C. Messenger, & F. Michon (Eds.), Decent working time: New trends, new issues. Geneva: International Labour Organization. Davidsson, L., & Naczyk, M. (2009). The ins and outs of dualisation: A literature review. University of Oxford: Recwowe working paper, 02/09. Dølvik, J.E., Fløtten, T., Hippe, J.M., & Jordfald, B. (2014). Den nordiske modellen mot 2030. Et nytt kapittel? Fafo-rapport 2014:46. Oslo: Fafo. Eldring, L., Ødegård, A. M., Andersen, R. K., Bråten, M., Nergaard, K., & Alsos, K. (2011). Evaluering av tiltak mot sosial dumping. Fafo-rapport 2011:09. Oslo: Fafo. Falkum, E., Hagen, I. M., & Trygstad, S. C. (2009). Bedriftsdemokratiets tilstand. Medbestemmelse, medvirkning og innflytelse i 2009. Fafo-rapport 2009:35. Oslo: Fafo Gallie, D. (2003). The quality of working life: Is Scandinavia different? European Sociological Review, 19(1), 61-79. Gallie, D. (Ed.) (2007). Employment regimes and the quality of work. Oxford: Oxford University Press. Hagen, I. M., & Trygstad, S. C. (2009). Local flexicurity: Resolving the conflict between direct and representative participation. Transfer 3-4/2009. Brussel: ETUI. Howell, C., & Given, R. K. (2011). Rethinking institutions and institutional change in European industrial relations. British Journal of Industrial Relations, 49(2), 231-255. ILO (1999). Decent work. Report of the Director-General to the International Labour Conference 87th Session. Geneva: ILO. Longva, F. (2001). Norway: Private sector services - A challenge to Centralised Co-ordination. In J. E. Dølvik (Ed.), At your service? Brussels: P.I.E. Lang S.A. Mahoney, J., & Thelen, K. (2010). Explaining institutional change: Ambiguity, agency, and power. Cambridge: Cambridge University Press. Marginson, P., & Sisson, K. (2004). European integration and industrial relations: Multilevel governance in the making. Basingstoke: Palgrave Macmillian. Melding til Stortinget 29 (2010-1011). Felles ansvar for et godt og anstendig arbeidsliv. NHO Service (2009). Profesjonelt renhold. Oslo: NHO Service. Nergaard, K. (2014). Social democratic capitalism. In A. Wilkinson, G. Wood, & R. Deeg (Eds.), The Oxford handbook of employment relations. Oxford: Oxford University Press. Nergaard, K., & Stokke, T. A. (2007). The puzzles of union density in Norway. Transfer, 4. Oliver, C. (1991). Strategic responses to institutional processes. The Academy of Management Review, 16(1), 145-179. O’Sullivan, M., & Turner, T. (2014). Protecting Precarious Workers in a Recession – Social Partners Strategies in the Face of Deregulation in Ireland. IREC Conference 2014 10-12 September, Dublin. Palier, B., & Thelen, K. (2010). Institutionalizing dualism: Complementarities and change in France and Germany. Politics & Society, 38(1), 119- a48. Pierson, P. (2000). Increasing returns, path dependence, and the study of politics. The American Political Science Review, 94(2), 251-267. Scott, R. (1995). Institutions and organizations. Los Angeles: Sage. Selznick, P. (1957). Leadership in administration. Evanston: Row, Peterson.

200 

 References

Sisson, K. (2013). Private sector employment relations in Western Europe. Collective bargaining under prssure? In J. Arrowsmith & V. Puligano (Eds.), The transformation of employment relations in Europe. London: Routledge. Sisson, K., & Marginson, P. (2001). “Soft regulation” – Travesty of the real thing or new dimension? Workingpaper 32/01, Sussex European Institute. Skilbrei, M-L. (2009). Fra tradisjonelt til moderne renhold: om kompetansekamper i renholdsbransjen. In K. Christensen& L. J. Syltevik (Eds.), Kvinners arbeid. Unipub forlag. Stokke, T. A. (2004). «Regulering i lov eller avtale». In NOU 2004:5 Arbeidslivslovutvalgets instilling. Et arbeidsliv for trygghet, inkludering, vekst. Thelen, K. (2012). Varieties of capitalism: Trajectories of liberalization and the new politics of social solidarity. Florence: Max Weber Programme, MWP – LS 2012/03. Thelen, K., & Kume, I. (2006). Coordination as a political problem in coordinated market economies. Governance, 19(1), 11-42. Tolbert, P. S., & Zucker, L. G. (1996). The institutionalization of institutional theory. In S. R. Clegg, C. Hardy, & W. R. Nord (Eds.), Handbook of organization studies (pp. 175-190). Los Angeles: Sage. Publication. Traxler, F. (2003). Bargaining, state regulation and the trajectories of industrial relations. European Journal of Industrial Relations, 9(2), 141-161. Traxler, F. S., Blaschke, S., & Kittel, B. (2001). National labour relations in internationalized markets: A comparative study of institutions, change, and performance. Oxford: Oxford University Press. Trygstad, S. C., Bråten, M., Nergaard, K., Alsos, K., & Ødegård, A. M. (2012). Vil tiltakene virke? Status i renholdsbransjen 2012. Fafo-rapport 2012:59. Oslo: Fafo Trygstad, S. C., Nergaard, K., Alsos, K., Berge, Ø. M., Bråten, M., & Ødegård, A. M. (2011). Til renholdets pris. Fafo-rapport 2011:18. Oslo: Fafo.

Torgeir Nyen, Anna Hagen Tønder

11 Cooperation and Reform in Vocational Education and Training In recent years national models of skill formation and vocational education and training have attracted attention from a large number of researchers. In the literature on varieties of capitalism (VoC), vocational education and training (VET) is one of the complementary institutions that distinguish liberal market economies (LMEs) from coordinated market economies (CMEs). LMEs, such as the United States, tend to encourage the acquisition of general skills. CMEs, such as Germany, place a stronger emphasis on vocational training as an alternative to or in addition to academic skills and higher education (Hall & Soskice, 2001). There is a broad consensus in the literature that different training regimes and the skills they promote have important economic and distributional effects. One limitation to this literature is that the categories provided are extremely broad and cannot account for variation within the different models. While all the Scandinavian countries can be classified as coordinated market economies, there are, as we will return to, major differences between the VET systems in Norway, Sweden and Denmark. In the comparative literature on national skill formation systems a number of more fine-grained typologies have been developed. The main dimensions that are applied in order to distinguish between different models are the structure and content of training, how the training is regulated, where the training takes place and the degree of involvement of the state and of firms in vocational education and training. One widely used typology distinguishes between market-based, statebased and corporatist training regimes (Greinert, 2004). A liberal market economy model is characterized by weak government regulations. Vocational training takes place mainly in the workplace and is provided by individual companies based on firm-specific skills demands. Britain and the United States often serve as examples. In a state-regulated bureaucratic model, vocational education is primarily governed by the state, whereas the social partners and private firms play a more limited role. The training is mainly school-based or mixed with shorter practice periods in firms. Training tends to focus on general academic and broad vocational subjects, with less emphasis on vocational specialization. More vocation-specific training is to a large extent developed in the workplace but is not part of the formal training. Sweden and France are typical examples of state-based regimes. In a dual corporate model, the social partners play a significant role in vocational education and training. Vocational education is formally regulated by the state and based on a cooperation model in which the social partners as well as educational authorities are involved. Firms play an important role in the provision of training. Apprenticeship training or practical training in the workplace is often combined with school-based training through some © 2015 Torgeir Nyen, Anna Hagen Tønder This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.

202 

 Cooperation and Reform in Vocational Education and Training

kind of dual system. Germany is the classic example, and Denmark is often mentioned as another country with a dual corporate model. In this chapter we focus on institutional change in Norwegian VET. Norway provides an interesting case for several reasons. With a few exceptions (e.g. Thelen, 2014), Norway is rarely mentioned in the literature. In her influential study of How Institutions Evolve, Kathleen Thelen studied the political economy of skills in Germany, Britain, the United States and Japan (Thelen, 2004). These countries are classic cases that are often related in comparative studies. A recent volume on the political economy of skill formation includes country studies from the Netherlands, Switzerland, Austria and Denmark in addition to Germany (Busemeyer & Trampusch, 2012). The Norwegian model is an interesting case because it can be characterized as a mixed model with elements from different skills regimes. After Reform 94, the Norwegian model can be classified as a hybrid between a state-based and a corporatist model (Olsen, Høst & Michelsen, 2008). One element from the statebased regime is that initial VET is deeply integrated in the national education system. School-based education and apprenticeship training are both regulated by national written curricula. Formal decision-making power on issues regarding the content and structure of the trades rests with the national education authorities. However, there is an established corporatist infrastructure of tripartite bodies with advisory functions at the national and regional levels. The two initial years of school-based education and training include a considerable share of general academic subjects like Norwegian, English, social science and mathematics. The content of school-based vocational courses is quite broad (with relevance for a number of different trades). The intention is to provide vocational students with broad and general skills in addition to the more vocation-specific skills. Norway has a unitary school system at the upper secondary level, with vocational programmes and general academic programmes offered within the same schools and with opportunities to combine vocational and academic training or to switch from a vocational programme to an academic programme. In these respects there are similarities between the Norwegian and the Swedish models. On the other hand, the role of apprenticeship training constitutes an important difference between the Norwegian and the Swedish models and brings the Norwegian VET model closer to the dual system in the Danish model than the school-based Swedish model. During the last two years of initial vocational education and training, students in the Norwegian model are apprentices employed by a training company. As apprentices, they receive wages that are regulated by collective agreements. The model is dual, both in terms of where the training takes place and in terms of a decision-making structure based on tripartite cooperation at the national and regional levels. Over the last decades VET as an institution has undergone substantial changes. A challenge to the VoC literature and to the different typologies of VET is to explain where the different models and institutions come from and how they change (Thelen, 2004). An analysis of institutional change in VET needs to deal with why institutions



Cooperation and Reform in Vocational Education and Training  

 203

sometimes are “reproduced” and why they sometimes change more profoundly. The history of VET in Norway serves as an example of how substantial changes do not necessarily require external shocks but can be endogenously motivated either by political reform or through more gradual change processes. To understand such changes, a historical-institutionalist analysis needs to study the actors’ interests and power bases, while at the same time acknowledging that existing institutions may influence the choices made by the actors involved (Thelen, 2004, 2010). An institution like a VET system is the result of power struggles, negotiations and compromises between individual employers, between different groups of employers, between employers and labour and between the social partners and the state. The system as such will be inherently unstable as actors seek to change the institutions, especially if changing external conditions also changes the strategic interest or power base of some of the actors. In the analysis that follows, we describe how a long-standing coalition between employers in crafts and manufacturing had a major impact on the development of the current dual model of VET in Norway. Until the 1950s, apprenticeship training constituted the core of vocational training in crafts and industry. State regulation was weak, and the social partners had great autonomy in their regulation of the training. The post-war period was characterized by stronger state involvement and the gradual development of a vocational school system. Still, the social partners had considerable autonomy in the development, implementation and control of apprenticeship training. In this period, Norwegian VET could be defined as a corporatist model. The scope of the model, however, was confined to craft and industry, and the number of new apprenticeship contracts that were signed every year was low. In the 1980s, new legislation expanded the scope of apprenticeship training and provided an institutional framework for tripartite cooperation in VET. The changes contributed to a revival of apprenticeship training in Norway. However, the connections between school-based education and apprenticeship training were weak. In the 1990s, Reform 94 drove the content of VET towards more general skills and broader vocational skills. The dual model was institutionalized as the standard model in all vocational programmes in upper secondary education, including sectors without strong traditions of apprenticeship training. The changes in content and structure of VET can be seen as a compromise solution that served the interests of employers in manufacturing and services more than the interests of employers within traditional crafts and construction. However, a further strengthening of apprenticeship training through policy measures, including increased state funding to training companies, made the employers within crafts and construction accept the reform. In recent years, tensions within the VET system have become more manifest. Changes to the current model have been proposed by several actors, among them the employers’ association within construction. Increased labour immigration has altered the veto power of employers in construction and in the service sector and is a factor that could generate changes that lead to greater heterogeneity in the Norwegian VET model in the future.

204 

 Cooperation and Reform in Vocational Education and Training

11.1 Historical Background: A Strong Coalition Between Employers in Crafts and Industry In Norway, like many other European countries, apprenticeship training initially developed within the guild system. In the 1800s, government policy was characterized by economic liberalization. A law passed in 1839 brought severe restrictions on the guilds, and new laws passed in 1866 and 1871 led to a complete abolition of the traditional guild system in Norway. With the new handicraft legislation, legal provisions for apprenticeship training and journeyman’s examinations were removed, meaning that in principle any man could practice a trade without skill certification. The liberalization of the handicraft sector led to a significant weakening of the apprenticeship system in the late 1800s (Kvikstad, 1998). After several unsuccessful attempts to re-establish regulations for apprenticeship training and trade examinations, the craft masters joined forces with employers in the rapidly expanding manufacturing sector. The Norwegian Association for Crafts and Industry (“Den Norske Fællesforening for Haandværk og Industri”) was established in 1886 and would become a major player in the development of vocational education and training in Norway. The alliance between crafts and industry in the early industrial period in Norway strengthened the influence of the employers as a strong collective actor. Proposed measures from the newly established Association for Crafts and Industry and the Crafts Association of Christiania1 (“Christiania Haandværkerforening”) had a substantial impact on a new handicraft act that was passed by Parliament in 1894. According to the new legislation a man who wanted to practice a trade regulated by the law had to pass a journeyman’s examination judged by masters within the relevant trade. The new legislation largely met the requirements that the craft masters and their associations had been trying to achieve for decades and provided a new foundation for apprenticeship training in Norway at the turn of the century. With the industrialization of the late 1800s, work within crafts and workshops gradually changed. Craft masters were increasingly seen as employers, whereas journeymen saw themselves as wage earners. The first trade unions and the first social democratic unions in Norway were formed by journeymen in Christiania (Kvikstad, 1998).

11.1.1 Apprenticeship Training in Industry Regulated in Collective Agreements In her comparative analysis of skill formation systems in Germany, Britain, the United States and Japan, Thelen (2004) documents that a decisive factor explaining the

1 Oslo was called Christiania (or Kristiania) from 1624-1924.



Historical Background: A Strong Coalition Between Employers in Crafts and Industry  

 205

different trajectories of skill formation was the strategies of leading firms in skillintensive industries and that the machine and metalworking industries were of special importance. As we will see in this section, these industries also had a major influence in the development of the Norwegian VET model through collective agreements and through the establishment of vocational schools. In the period from 1890 to 1915 the number of industrial workers in Norway more than doubled (Fuglum, 1978). The industrialization process had a strong impact on the employers’ willingness to invest in apprenticeship training. More routine work meant that many tasks could be performed by unskilled workers. The tendency to displace apprentices and skilled workers with unskilled or semi-skilled workers was particularly evident in the metalworking industry. The workers’ interest in protecting the position and wage conditions of skilled workers was a driving force behind the first nationwide collective agreements in Norway. The first agreement was made within the metal manufacturing industry in 1907 (“Jernavtalen”). These agreements established minimum wages according to skills and made it less profitable for employers to rely on unskilled workers instead of investing in skills and training (Andersen, 1984). Furthermore, the agreements established the mutual recognition of the organized parties (Løken & Stokke, 2009). The traditional crafts had relied on skills that could be developed in the workplace through close observation and imitation of the master’s work. Most of the work was performed manually, and the firms were typically small, with only one or a few apprentices. Manufacturing firms were usually larger, work was more mechanized and there could be a large number of apprentices. Some of the skills needed by industry workers could be developed more effectively in schools, separate from the production process in the workplace. From the early 1900s a number of vocational schools was established, where young people were trained before entering into apprenticeship training in the workplace. The state initially played no active role in the establishment or development of these schools. Employers, especially within the metalworking industry, were the main driving forces, and the Association for Crafts and Industry was a key player in the development of the vocational school system. Traditionally, the handicraft masters had defended their autonomy from the government in matters concerning apprenticeship training. Within the new alliance between crafts and industry, this strategy gradually changed. The employers argued that when part of the training took place in schools and provided the students with vocational skills that would be useful to many employers, the state should contribute financially. At the same time, employers wanted to have control over the training content (Bjørndal, 2005). This ambivalence towards state involvement among the employers has been characteristic in the development of VET. Several training schools for crafts and industry were established in Kristiania in the early 1900s, mainly initiated by employers and their organizations. The vocational training school for metal workers was established in 1910 and formed the pattern for a number of other schools. The training school for metal workers was established

206 

 Cooperation and Reform in Vocational Education and Training

with financial support from the state, the municipality and a number of private metalworking companies, such as Aker Mechanical Workshop, Myrens Workshop and Thunes Mechanical Workshop. The purpose of the school was to provide young men with a basic knowledge of tools, materials and methods and with some basic vocational skills before entering apprenticeship training in a workshop (Andersen, 1984). Until World War II the development of vocational education and training in Norway was to a large extent driven from “below” on the basis of initiatives from individual employers and organizations out of the perceived need for skills within crafts and later from manufacturing. A multitude of different training programmes and vocational schools were established without an overall plan or systematic policy. As we have seen, workshop schools or vocational schools providing initial vocational training were established at the beginning of the 1900s. However, the development of the school system was slow, and few schools were established outside the city of Oslo. Apprenticeship training and informal training in the workplace still constituted the core in the development of skilled workers in crafts and manufacturing.

11.1.2 The Post-War Period: The State Takes a More Active Role While vocational education and training had previously been characterized by great diversity and weak state regulation, the post-war period was marked by institutional development and attempts to standardize vocational training. Gradually, a corporatist training system developed within crafts and industry. The social democratic government elected in 1935 had an ambition to develop a more modern and better organized system for vocational education and training. The field of VET became increasingly politicized and related to social and economic issues. The School for Trades and Industry Act was passed in 1940 and came into force in 1945, after the war. According to the new law, vocational schools should be developed into a nationwide system. At this time, the view that school-based vocational education should precede apprenticeship training had become more influential, but was still not the dominant view among employers (Skule, Stuart & Nyen, 2002). The Apprenticeship Act was introduced in 1952 and regulated many aspects of apprenticeship training. Employers that would train apprentices in trades covered by the law were required to sign apprenticeship contracts and to take on clear commitments in terms of training. In places where an apprentice school had been established, the apprentice should attend these schools and the employer should provide the apprentices with time off for school work. Written curricula were developed for training in the trades regulated by the law. An administrative structure, with representatives from the social partners at the national and regional levels, was developed. Within this structure, the social partners were provided with considerable autonomy in the development, implementation, evaluation and control of the



Historical Background: A Strong Coalition Between Employers in Crafts and Industry  

 207

apprenticeship system (Høst, 2008, p. 87). The scope of the law gradually increased. In 1952, the Apprenticeship Act comprised only six trades within traditional crafts. By 1975, the number of trades comprised by the law had increased to 149 (NOU 1976: 10). The Apprenticeship Act was the first major step in an attempt by the Norwegian state authorities to regulate and control vocational training in the workplace. These attempts were met with ambivalence or direct opposition among employers. One of the elements that made it easier to accept the new regulations was the establishment of an alternative route to obtain a trade certificate based on work experience. Adults who have acquired vocational skills through work experience can register for a theoretical and practical trade examination as so-called practice candidates. This institutional arrangement has existed since the regulation of apprenticeship in the 1950s and still plays an important role in the Norwegian VET system. Within some industries, the practice candidate arrangement is still the most important way to formal skills certification. In the 1950s the vocational school system in Norway was further developed and expanded. The intention was that initial training in vocational schools should be followed by apprenticeship training in a firm and the training be completed with a trade or journeyman’s certificate. In practice it turned out to be difficult to obtain a good match between the number of students in school-based vocational training and the number of available apprenticeships. Whereas the vocational school system expanded rapidly in this period, the number of apprenticeships remained at the same level as before (Olsen, Høst & Michelsen, 2008). Another issue that again was being debated was the content of school-based vocational training. The school system that had developed over time was characterized by a great diversity and different educational traditions in areas such as crafts, manufacturing, sales and administration, shipping and health care. The educational authorities gradually saw a need to systematize a school system that was highly fragmented. An important political aim was to offer equal educational opportunities to students regardless of where they lived. A government committee was appointed in 1965 to prepare a new legislative framework for upper secondary education (“Steenkomiteen”). The committee proposed to combine vocational programmes and general academic programmes in a unitary upper secondary school system. According to the committee, the modernization and structural changes of working life would require more theoretical skills and a greater emphasis on general academic subjects in vocational training. An important objective was to ensure a broader and more democratic recruitment to higher education. The Act of Upper Secondary Education was passed in 1974, removing the formal division between general and vocational education at the upper secondary level. Since then, most upper secondary schools in Norway have offered both general and vocational programmes. The Reform of 1974 was confined to the upper secondary school system and did not include apprenticeship training.

208 

 Cooperation and Reform in Vocational Education and Training

In the 1970s apprenticeship training had a weak position in Norway, with only 2-3000 new apprenticeship contracts signed every year. Some actors, including the youth organization of the Labour Party (AUF), were in favour of abolishing apprenticeship training altogether. The main argument was that the apprenticeship system maintained and reinforced social inequalities. The opponents of apprenticeship training argued for promoting social mobility through general education and access to higher education: privileges that had previously been reserved for the middle class (Mjelde, 2002; Høst & Hovdhaugen, 2013). Nevertheless, the apprenticeship system survived as a recruitment and skills development system outside or regardless of the ongoing integration process in upper secondary education. Collective actors within crafts and manufacturing continued their efforts to maintain a connection between school-based vocational education and the occupational categories in the workplace. The Act of Upper Secondary Education from 1974 established a unitary school system at the upper secondary level, but the vocational programmes and traditions that had developed within the vocational schools continued to exist within the new institutional framework. The result was a very heterogeneous structure with a wide variety of courses of different length and composition (Høst & Hovdhaugen, 2013).

11.1.3 The 1980s: Revitalization of Apprenticeship Training In the 1980s interest in the apprenticeship system gained new momentum. Legislative measures as well as new cooperative measures between firms contributed to the expansion and renewal of the apprenticeship system as a training model. In 1981 the Apprenticeship Act of 1950 was replaced by the new Act on Vocational Training. A main objective was to strengthen the workplace as a training venue. The new act provided an institutional framework for close cooperation between the government and the social partners, thereby establishing a model of tripartite cooperation in Norwegian VET. The social partners were provided a strong influence on the content of the training in order to ensure that curricula were consistent with the needs of working life, whether the training took place in schools or in workplaces. While the Apprenticeship Act only regulated apprenticeship training in cities and urban areas, the new Act on Vocational Training applied to the entire nation. The legislation was also more extensive in the sense that new sectors and trades were comprised by the law. The new trades included industry trades within chemical processing, food processing and lumber manufacturing (Høst, Gitlesen & Michelsen, 2008, p. 22). The establishment of new trades made it possible to upgrade work from semi-skilled to skilled work, a factor that was clearly in the interests of labour, as it provided for higher wages. In the middle of the 1980s the number of new apprenticeship contracts had increased from 2-3000 per year to around 10 000. The new legislation and institutional framework and the expansion of the legislation to



Historical Background: A Strong Coalition Between Employers in Crafts and Industry  

 209

new trades and new geographical areas were important factors that contributed to the strengthening of the apprenticeship system during this period. Another measure that probably had a positive impact on the number of apprenticeships was an increase in the state grant given to training companies for each apprentice. In addition to the policy measures, a number of new training offices (“opplæringskontor”) were established by employers in order to strengthen cooperation in apprenticeship training. The training offices are collective enterprises that are owned by training companies. The apprenticeship contracts are signed by the training offices. The training offices assist the training companies in their training of apprentices, thereby dividing responsibilities and reducing the training costs of individual firms. They also work actively to recruit new training companies and to find apprenticeship places among the existing members (Bjørndal, 2005; Hagen & Skule, 2007; Kuczera et al., 2008). Even if there was a strong increase in the number of apprenticeship contracts in the 1980s, the number of apprenticeships was still heavily influenced by economic cycles (Høst, Gitlesen & Michelsen, 2008). This became particularly evident towards the end of the 1980s when the economic downturn dramatically altered the situation of the labour market. The development in the youth labour market was an important backdrop for Reform 94 as an extensive reform of upper secondary education and the Norwegian VET system.

11.1.4 Reform 94: Apprenticeship Training an Integral Part of Upper Secondary Education The youth labour market in Norway was severely affected by the decline in manufacturing employment in the 1970s and 1980s. In 1992, the number of teenagers employed in manufacturing had been halved since 1980. At the beginning of the 1990s total employment among the youth had declined by nearly 30 per cent since the 1970s (NOU 1994:3). As a consequence of reduced labour market opportunities, an increasing number of youths and young adults applied to upper secondary schools. The counties, who are responsible for upper secondary education in Norway, did not have enough room in the schools, meaning that a large number of applicants had to be turned down. Every year, state funding to the counties was increased so that they could increase their capacity and take in more students, causing the most significant expansion of upper secondary education in Norwegian history. In 1980 there were around 137 000 students in upper secondary education. Ten years later, the number had increased to more than 200 000 students. A low progression rate in vocational programmes reinforced the capacity problems in upper secondary schools. Many students left school without formal qualifications or they moved horizontally from one basic course to another without moving on in the system. The school structure, with a high degree of vocational specialization starting

210 

 Cooperation and Reform in Vocational Education and Training

from the first year, was considered a main problem by the education authorities. A political goal for the social democratic government was to grant all youth a statutory right to upper secondary education. In 1989, a committee was appointed (the Blegen Committee), which included representatives from LO and NHO, to prepare a report on how this political goal could be achieved (NOU 1991:4). The work of the committee stimulated discussions between the social partners about how to secure the provision of skilled labour and of vocational training in the future. These discussions resulted in the document “Joint Declaration on vocational education and training in schools and workplaces”, signed by The Norwegian Confederation of Trade Unions (LO) and the Confederation of Norwegian Enterprise (NHO). The document has been described as a cornerstone in the development of the reform (Bjørndal, 2005). The agreement between LO and NHO in the joint declaration assumed that the basic theoretical training should take place in upper secondary schools. The basic courses in vocational programmes were to provide a broad basis for further training with an increased emphasis on general subjects such as Norwegian, English and mathematics. The idea was that a broader basic education would give vocational students a better foundation for lifelong learning and skills in line with future needs within the trades. Vocational specialization and practical training in the trade should primarily be the responsibility of the workplaces in the second phase of the training programme. The social partners should work to increase the number of apprenticeship places so that all youth who applied to a vocational programme would be given the opportunity to complete their education. The main features of Reform 94 were based on the report from the Blegen Committee and the Joint Declaration of the social partners. The so-called 2+2 model (two years of school-based education followed by two years of apprenticeship training) was established as the main model for all vocational education and training. Through Reform 94, the apprenticeship system was formally integrated as a part of upper secondary education, and vocational education and training became the shared responsibility of the state and the social partners. The content of vocational education and training and the balance between breadth and depth have changed over time. Changes in the content and composition of vocational programmes could be related to changes in the institutional foundation of VET and in the changing roles and responsibilities of the state and the social partners (Nyen & Tønder, 2013). The establishment of the 2+2 model as the standard model can be seen as a compromise solution, seeking to balance the need for general education and breadth in vocational programmes with the need for vocational specialization and practical training in the workplace. Furthermore, the standardization of the dual model established a clear institutional division between schools and workplaces in the first and second part of the training programme, respectively.



Historical Background: A Strong Coalition Between Employers in Crafts and Industry  

 211

11.1.5 Reform 06: Broader Courses with Opportunities for Specialization In 2006 a new reform, named the “Knowledge Promotion Reform” was introduced. With this reform, new changes in the structure were implemented with fewer and broader vocational courses in the first and second year of vocational education and training. Still, the 2+2 model was retained as the main model in vocational training. The white paper “Culture for Learning” emphasized that the new programme structure was introduced in order to reduce problems associated with low progression in upper secondary education and to facilitate recruitment to small trades and occupations. Furthermore, it was argued that broader courses would ensure that students could make their educational choices regardless of geography and financial status. For students who were still undecided on their choice of education, broader “entrances” to vocational programmes would make it easier to postpone the choice of education, thereby reducing the risk of wrong choices and delays in education. At the same time, it was emphasized that students who had decided on a specific career before entering secondary education should still have the opportunity to practice their preferred occupation at an early stage in their training. The white paper also introduced the concept of the “In depth Study project” (PTF). The introduction of PTF is specifically related to the need for vocational specialization. It is argued that students who are interested in particular occupations should have an opportunity to get in contact with and practice these occupations at an early stage in their training. If a school cannot offer the specialization wanted by a student, the student should be given the opportunity to receive parts of their training in a workplace or at another school. A central assumption is that the opportunity to be introduced to a specific occupation at an early stage will increase student motivation and learning. However, the structural changes and the introduction of PTF in the reform are not only justified in terms of student needs. The white paper also argued that the combination of broader programmes and opportunities for vocational specialization is necessary in order to accommodate skills needs in different parts of the labour market (Nyen & Tønder, 2013).

11.1.6 Recent Developments Companies and employer organizations in some parts of the labour market have expressed strong concerns that trade-specific competence is not sufficiently developed within the current structure. This leaves companies with a de facto greater responsibility in developing vocational skills within the two years of apprenticeship training. There are clear signs of discontent with the competence level of new apprentices after the introduction of Reform 94 and Reform 06 in traditional apprenticeship trades, like the construction industry.

212 

 Cooperation and Reform in Vocational Education and Training

In 2012, The Federation of Norwegian Construction Industries, Byggenæringens landsforening, or BNL, proposed changes in the training model as well as in the decision-making rules for the national vocational training council within their sector. BNL argued for an “alternating model”, a model with more frequent changes between learning at school and learning in the workplace, allowing for earlier trade specialization. In BNL’s view, such a model would strike a better balance between broad and trade-specific competence. BNL also called for decision-making rules giving the employer and employee organizations a more decisive influence on the vocational training system. Some schools have been practicing 1+3 models (one year in school, three years of apprenticeship). Others are currently testing the alternating model suggested by BNL. In other words, there are signs of institutional change in terms of layering – new practices are tried out along with the established 2+2 model. In other parts of the labour market, there has been less discontent with the outcomes of the Knowledge Promotion Reform. For instance in health and care services and kindergartens, municipal employers were generally pleased with the new health worker trade as well as the childcare and youth work trade (Nyen, Reegård & Tønder, 2011). However, these trades face a number of other challenges, not least a growing number of students leaving the VET programme after the second year, bridging across to the general programme in order to qualify for higher education (Skålholt, Høst, Nyen & Tønder, 2013). Formally, the 2+2 model is still the main model within these trades. In practice, however, there are strong tendencies towards academic drift, gradually undermining the role of VET in the service sector. In March 2013, a new government white paper, St.meld. 20 (2012-2013) På rett vei (“On the right track”), was presented to Parliament. The paper signalled changes allowing for greater variation in vocational programmes in order to meet diverse needs in different parts of the labour market. The white paper also encouraged stronger emphasis on trade specific competence in the VET programmes and called for an evaluation of the various VET trades by the labour market parties and national education authorities. Other measures include the introduction of a statutory right for vocational students to an extra year at school after they have obtained a trade certificate in order to gain entry qualifications for higher education. The measures signalled in the white paper introduce more flexibility and allow greater heterogeneity between the education programmes and trades at the national level. It encourages new discussions and negotiations between the state and the social partners at the national level about the content, structure and role of VET in each sector and industry. An institutional infrastructure for such negotiations has already been established within the national trade councils but may be given more impetus. To some extent, there is a process of layering going on, as new models like the alternating model or the vekslingsmodell are allowed without replacing or altering the basic structure of subjects and the main model of 2+2. More importantly, the white paper allows for greater adaptation of competence and training models to the needs expressed by the employers’ organizations.



Historical Background: A Strong Coalition Between Employers in Crafts and Industry  

 213

11.1.7 Institutional Changes Our aim in this chapter has been to identify the main factors driving the development of the vocational education and training in Norway over the last century. We have emphasized the implementation of Reform 94, with the establishment of the dual model as a standard model of training in all vocational programmes in upper secondary education. Before this quite substantial reform, Norwegian VET could be described as a corporatist model, where the social partners had considerable autonomy in the development and implementation of the system. The main actors among the social partners in the preparation of the reform were LO and NHO, continuing a collaboration that had developed over a century. The position of the trade unions and the cooperation between employers’ organizations and trade unions in Norway was an important factor in the development of vocational education and training. The distinction between skilled and unskilled workers has been a key element in collective agreements in Norway since the first nationwide collective agreement was signed within the metalworking industry in 1907. These agreements had a major impact on the development of a cooperative system of industrial relations in Norway, including the regulation of skills and training (Løken & Stokke, 2009). Employers in the machine and metalworking industry were important actors in the establishment of a vocational school system separate from production. Around the school issues, the interests of crafts and manufacturing were partly divergent. The demand for separate schools was mainly in the interest of the employers in manufacturing. Within the crafts, the preferred model was to develop theoretical as well as practical skills in the workplace, closely integrated with the actual production process. The handicraft masters preferred to have autonomy from the state and to keep close control over the training process and the development of their own trade. This divergence in interests combined with the close coalition between crafts and manufacturing has been resolved through negotiations, conflict and compromise throughout the history of VET. The tripartite cooperation between the state and the social partners is another essential factor when trying to understand the development of the Norwegian VET system in general and the institutionalization of a dual model through Reform 94 in particular. Before the 1950s state involvement in vocational education and training was weak. Apprenticeship training in traditional crafts was to some extent regulated in handicraft legislation. Industry trades were not regulated by legislation, but wage provisions for apprentices and skilled workers were regulated by collective agreements. To the extent that the typologies of different training regimes are applicable to this period, we would say that vocational education and training in crafts and industries could be characterized as somewhere between a corporatist regime and a marketbased regime. Changes within the system before the 1950s were small and gradual.

214 

 Cooperation and Reform in Vocational Education and Training

The Apprenticeship Act was the first major step in an attempt by the state authorities to regulate apprenticeship training across industries. The regulation introduced elements from a state-based model, but the social partners were still granted a high level of autonomy in the administration and control of the apprenticeship system. In addition, the establishment of the experience-based trade certificate scheme alongside the apprenticeship system meant that it was possible to bypass the apprenticeship training regulations and train skilled workers through informal learning in the workplace. The existence of two alternative routes to the trade certificate can be seen as an example of layering in the sense that old practices could continue alongside new legislation (Mahoney & Thelen, 2010). The replacement of the Apprenticeship Act by the new Act on Vocational Education and Training in 1980 strengthened the corporative elements of the Norwegian model and institutionalized tripartite cooperation in vocational education and training at the national and regional levels. In a comparative perspective, the delegation of responsibilities and power to the social partners within a tripartite framework was quite remarkable (Olofsson, 2010, p. 223). The revision of the Apprenticeship legislation was partly initiated by the social partners. The new legislation was made applicable to the whole country and included a number of new trades. In this sense, the changes could be seen as an expansion and a strengthening of an existing model more than as a replacement of an old model with a new one. Reform 94 is the most distinct institutional change process in the history of Norwegian VET so far. The reform integrated apprenticeships and school-based vocational education into a unified standard model for upper secondary vocational education. The social democratic government was under heavy pressure with record high youth unemployment coupled with capacity problems in upper secondary schools. The fragmented structure in vocational programmes contributed to the problem with low progression among vocational students. The challenge was to find a way to offer all youth a statutory right to upper secondary education. The apprenticeship system offered a solution to the problem, and the 2+2 model became the standard model for all vocational programmes, including industries outside crafts and industry, like health care and the private service sector. The longstanding alliance between crafts and manufacturing was instrumental in obtaining employer support for the reform. Employers in both crafts and manufacturing were organized within the confederation of Norwegian enterprises, NHO, which allowed for intra-organizational bargaining with a degree of mutual trust. Although Reform 94 served the interests of manufacturing more than crafts, all parties had a common interest in designing a system that attracted youths to VET and to industries that recruited through VET. Opting out of the VET system was not a viable solution for crafts, which historically has been the core segment for VET and apprenticeships. The balance between breadth and depth and the timing of specialization are issues that have always been and that still are debated. It could be argued that the traditional crafts were the main losers in Reform 94 and Reform 06, with the introduction of



Historical Background: A Strong Coalition Between Employers in Crafts and Industry  

 215

more general academic subjects, broader vocational programmes and less vocational specialization. The tensions and unresolved conflicts from earlier compromises have become more manifest in recent years as employers’ organizations within traditional crafts have expressed their discontent and demanded changes in the content and composition of vocational training and education. The liberalization of labour markets has provided these employers with alternative options for recruitment and training, thereby increasing their veto power. The classic dilemma to be resolved is that recruitment strategies that are profitable for the individual employer in a shortterm perspective may lead to an erosion of skills and outcomes that are unfavourable to everyone in the long run. In order to deal with these dilemmas, new solutions will constantly have to be developed between employers, labour and the state. The history of Norwegian VET has not been a linear process. At certain stages developments could have taken a different turn. The coalition of manufacturing and crafts helped to establish a dual VET model for training and recruitment within these sectors, where apprenticeships were gradually combined with school-based education through the first half of the 20th century. Unlike in Sweden, apprenticeships and the dual model survived the increased state involvement and broadening of general education in the 1950s and 1960s. In Sweden a lack of cooperation between companies and between labour and employers, combined with a negative position of organized labour to apprenticeships, led the labour market parties to leave initial VET to the state (Olofsson, 2005). In Norway, apprenticeships survived among other things because organized labour accepted wage differentials between skilled and unskilled labour and because of a degree of cooperation and trust between employers. Still, the number of apprenticeships was low in the 1960s and 1970s. The resurgence of the dual VET model through Reform 94 demonstrated the capacity of the tripartite system to reform and change VET policies. It remains to be seen whether the tripartite deliberative system will be able to deal with the current challenges to VET, not least because solutions must be found across various policy fields. However, recent history shows that there has been a mutual trust and willingness among the labour market parties and government to find solutions to these challenges.

216 

 References

References Andersen, E. (1984). Norsk yrkesopplæring 1890-1940. Hovedoppgave. Oslo: Statens yrkespedagogiske høyskole. Bjørndal (2005). Videregående opplæring i Norge i 800 år – med hovedvekt på tiden etter 1950. Halden: Forum Bok. Busemeyer, M. R., & Trampusch, C. (2012). Introduction. The comparative political economy of collective skill formation. In M. R. Busemeyer & C. Trampusch (Eds.), The political economy of collective skill formation (pp. 3-40). Oxford: Oxford University Press. Fuglum, P. (1978). Norge i støpeskjeen 1884-1920. Norges historie, volume 12. Oslo: Cappelen. Greinert, W.-D. (2004). European vocational training systems: The theoretical context of historical development. In Cedefop, Towards a history of vocational education and training (VET) in Europe in a comparative perspective (pp. 17-27). Luxembourg: Cedefop Panorama series 103. Hagen, A., & Skule, S. (2007). Den norske modellen og utviklingen av kunnskapssamfunnet. I J.E. Dølvik et al. (Eds.), Hamskifte. Den norske modellen i endring (pp. 145-168). Oslo: Gyldendal Akademisk. Hall, P. A., & Soskice, D. (2011). An introduction to varieties of capitalism. In P. A. Hall & D. Soskice (Eds.), Varieties of capitalism: The institutional foundations of comparative advantage (pp. 1-68). Oxford: Oxford University Press. Høst, H. (2008). Governance and cooperation. In H. Høst (Ed.), Continuity and change in Norwegian vocational education and training (VET) (pp. 87-94). NIFU STEP Rapport 29/2008. Oslo: NIFU STEP. Høst, H., Gitlesen, J. P., & Michelsen, S. (2008). Læreplasser mellom politikk og konjunkturer. In H. Høst (Ed.), Fag- og yrkesopplæringen i Norge – noen sentrale utviklingstrekk (pp. 17-28). NIFU STEP Rapport 20/2008. Oslo: NIFU STEP. Høst, H., & Hovdhaugen, E. (2013). Ny struktur – tradisjonelle mønstre. Kunnskapsløftets endringer i et historisk perspektiv. In B. Karseth, J. Møller & P. Aasen (Eds.), Reformtakter. Om fornyelse og stabilitet i grunnopplæringen (pp. 61-82). Oslo: Universitetsforlaget. Kuczera, M., Brunello, G., Field, S., & Hoffmann, N. (2008). Learning for jobs. OECD reviews of vocational education and training. Norway. Paris: OECD. Kvikstad, J. (1998). Arbeidsliv og fagopplæring før 1900. MA thesis, history. Trondheim: Norges teknisk-naturvitenskapelige universitet. Løken, E., & Stokke, T. A. (2009). Labour relations in Norway. Fafo-report 2009:33. Oslo: Fafo. Mahoney, J., & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change. Ambiguity, agency, and power (pp. 1-37). Cambridge: Cambridge University Press. Mjelde, L. (2002). Yrkenes pedagogikk, fra arbeid til læring – fra læring til arbeid. Oslo: Yrkeslitteratur. NOU 1976: 10. Framtidig yrkesutdanning. Samarbeid mellom skole og arbeidsliv. Oslo: Norges Offentlige Utredninger. NOU 1991: 4. Veien videre til studie- og yrkeskompetanse for alle. Oslo: Norges Offentlige Utredninger. NOU 1994: 3. Ungdom, lønn og arbeidsledighet. Oslo: Norges Offentlige Utredninger. Nyen, T., Reegård, K., & Tønder, A. H. (2011). Har vi lærefag som er relevante for kommunesektoren? Fafo-rapport 2011:36. Oslo: Fafo. Nyen, T., & Tønder, A. H. (2013). Prosjekt til fordypning – et taktskifte i fagopplæringen. In B. Karseth, J. Møller, & P. Aasen (Eds.), Reformtakter. Om fornyelse og stabilitet i grunnopplæringen (pp. 83-98). Oslo: Universitetsforlaget.

References 

 217

Olofsson, J. (2005). Svensk yrkesutbildning. Vägval i internationell belysning. Stockholm: SNS Förlag. Olofsson, J. (2010). Nordiska erfarenheter av lärlingsutbildning. In SOU 2010: 19, Lärling – en bro mellan skola och arbetsliv. Bilaga 1 (pp. 193-227). Stockholm: Sveriges Offentlige Utredningar (SOU). Olsen, O. J. (2008). Institusjonelle endringsprosesser i norsk- fag- og yrkesutdanning. Fornyelse eller gradvis omdannnelse? Notat 5-2008. Bergen: Rokkansenteret. Olsen, O. J., Høst, H., & Michelsen, S. (2008). Veier fra yrkesopplæring til arbeidsliv. En studie av det norske overgangsregimets effektivitet. I J. Olofsson & A. Panican (Eds.), Ungdomars väg från skola till arbetsliv. Nordiska erfarenheter (pp. 249-332). TemaNord 2008:584. Copenhagen: Nordisk Ministerråd. Skule, S., Stuart, M., & Nyen, T. (2002). Training and development in Norway. International Journal of Training and Development, 6(4), 263-276. Skålholt, A., Høst, H., Nyen, T., & Tønder, A. H. (2013). Å bli helsefagarbeider. En kvalitativ undersøkelse av overganger mellom skole og læretid, og mellom læretid og arbeidsliv i helsearbeiderfaget. NIFU STEP Rapport 5/2013. Fafo-rapport 2013:05. Oslo: NIFU STEP & Oslo: Fafo. Thelen, K. (2004). How institutions evolve. The political economy of skills in Germany, Britain, the United States and Japan. Cambridge: Cambridge University Press. Thelen, K. (2010). Beyond comparative statics: Historical institutional approaches to stability and change in the political economy of labor. In G. Morgan, J. L. Campbell, C. Crouch, O. K. Pedersen & R. Whitley (Eds.), The Oxford handbook of comparative institutional analysis (pp. 42-61). Oxford: Oxford University Press. Thelen, K. (2014). Varieties of liberalization and the new politics of social solidarity. Cambridge: Cambridge University Press.

PART III: Negotiations in Welfare State Institutions

Anniken Hagelund, Axel West Pedersen

12 To Reform or Not to Reform? Explaining the Coexistence of Successful Pension Reform and Sick Pay Inertia in Norway 12.1 Introduction Why do some aspects of welfare policy lend themselves to radical reform, while others seem to resist all demands for reform and retrenchment? How can we understand that institutional change takes place to such different degrees and in such different manners in different parts of the same welfare state? Differential reform capacity in different policy areas of the welfare state is a largely unexplored area of research. Yet, there are several advantages to comparing the occurrence of institutional change in different areas of the same welfare state. One is that such within-country comparative analysis allows us to “control for” the pressures of globalization and budgetary constraints that tend to dominate much of the literature and ask questions about other and more internal drivers of change. By focusing specific programmes and reforms we are able to examine the details of interest group configurations, institutional heritages and ideological struggles that pertain to specific policy areas and specific reform processes. In this chapter we try to explain what appears to be a puzzling difference between the contemporary dynamics of pension policy and sick pay policy in Norway. Despite persistent concerns about comparatively high sickness absence rates, the basic structure of the extremely generous Norwegian sickness insurance has remained nearly unchanged since 1978. This is all the more surprising as there have been several failed initiatives by successive governments to cut benefits and to improve the incentive structure vis-à-vis both employers and individual employees and because the case deviates from general tendencies towards retrenchment in other countries’ sickness insurance systems. The story of the Norwegian pension reform is quite different. Here the Norwegian political system in 2009 delivered a fairly radical structural reform that promises to secure economic sustainability in the face of an ageing population. The successful implementation of the pension reform is surprising in its own right as it has been enacted despite comparatively weak reform pressures and in the absence of an acute economic crisis that might otherwise have offered reform eager politicians an opportunity for breaking popular resistance (Pierson, 1996). In the following we first present three different theoretical perspectives which have been dominant in the literature on social policy reform processes. The first perspective views policy reform processes as the result of strategic games, emphasizing the preferences, priorities and power-resources of collective actors and stakeholders. The second perspective concentrates on institutional factors and mechanisms of path © 2015 Anniken Hagelund, Axel West Pedersen This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Theoretical Perspectives on Social Policy Reform 

 221

dependency. Finally, the third perspective focuses on the significance of discourse and framing in the policy reform process. We argue that there are good reasons to combine such perspectives when analysing complex policy processes, in particular as, in practice, they are not easily separated. Institutional configurations, such as the tripartite collaboration on policy making that is crucial in the field of Norwegian sickness insurance, cannot be understood without taking into account the strategic interests of the major players, but they also give shape to a particular kind of discourse with repercussions on the scope for future policy making. After introducing the theoretical perspectives we present the two cases briefly, before we move on to comparatively analysing the cases in light of the three perspectives.

12.2 Theoretical Perspectives on Social Policy Reform Power resource theory was for many years the dominant perspective in comparative welfare state research, particularly in the quantitatively oriented branch. According to Paul Pierson, this perspective “has had considerable success in accounting for cross-national variations in social provision during the three decades following World War II” (Pierson, 1996, p. 150). The essence of the power resource perspective is well captured by its leading proponent Walter Korpi (1989, p. 313): “…in the power resources perspective, political parties and other interest organizations (…) tend to appear as the organized expressions of collective action geared toward safeguarding the diverging interests of groups differently located in the social structure”. In comparative research this very general perspective was most often given a more narrow interpretation in which the conflict between labour and capital took centre stage and the mobilization of the working class through trade unions and left parties was seen as the key to welfare state expansion (Korpi, 1981). When Paul Pierson launched his famous thesis about a New Politics of Welfare, one of the central premises was that organized labour had been seriously weakened by the late 1980s and that the apparent resilience of developed welfare states to efforts at retrenchment must be explained by other factors (Pierson, 1996). Of course Pierson’s general claim on this point has been controversial. Particularly in Scandinavia the underlying premise about a serious weakening of organized labour as a political force does not seem to fit reality. In the Norwegian case it is rather obvious that any attempt to explain reform and resilience in welfare policy needs to take account of the positions taken by the trade union movement and its strategic interaction with the employers and the political authorities. Institutional approaches are concerned with the impact of welfare state structures and political institutions on the potential and direction of policy reform. While much early welfare state research interpreted the development of welfare state programmes as responses to functional needs or power constellations between social actors, institutionalism brought politics and the state “back in” (Skocpol, 1985).

222 

 To Reform or Not to Reform?

Much institutionalist writing focuses on characteristics of the political system and examines how these can impact the extent and shape of welfare policies. In these studies elements such as the electoral system, whether there is a president or a parliamentary system and the presence of veto points such as second chambers are crucial variables, sometimes in combination with partisan variables. However, for the comparative analysis we are doing of two different branches of welfare policy within the same country, these institutional political variables are less relevant – our cases are played out within the very same political system and in much of the same time period. However, de facto veto points may be differently set up in sickness insurance and pension policy as these are also embedded within the system of tripartite relations (including negotiated wage agreements). More relevant for our purposes are the bodies of literature that focus on the welfare state institutions themselves in shaping the conditions of reform. This type of analysis has tended to focus on feedback effects and path dependence. More specifically, we shall here examine what Pierson (2000, 2004) refers to as the logic of increasing returns. While path dependency – in its broader sense – simply refers to the causal relevance of preceding events (history matters), a narrower and theoretically more powerful understanding of the concept points to mechanisms whereby policy developments after some critical turning point become locked into a particular, self-reinforcing development track. This conception of path dependence is closely related to the idea of increasing returns, suggesting that there is a cumulative effect: “In an increasing returns process, the probability of further steps along the same path increases with each move down that path” (Pierson, 2000, p. 252). The emphasis on path dependencies naturally leads to a bias towards analyses of barriers to change, institutional inertia and stability. Against this, institutionalist scholars have more recently turned towards theorizing and studying gradual or incremental institutional change. Exogenous shocks may be necessary to generate radical institutional reconfigurations, but endogenous developments of institutions may still take place in more gradual and slow-moving ways. Thelen and co-authors suggest a typology of such incremental institutional change, distinguishing between displacement, layering, drift and conversion. In displacement existing rules are removed and new ones introduced. Layering involves the introduction of new rules on top of or alongside existing ones. Drift is the changed impact of existing rules due to shifts in their environment. Conversion means that rules remain formally the same, but are interpreted and enacted in new ways (Streeck & Thelen, 2005; Mahoney & Thelen, 2010). This perspective allows for greater sensitivity to the different kinds of political dynamics and actor constellations that underlie processes of change. Another twist to institutionalist writing is made by authors such as Vivien Schmidt (2002, 2008), who has pointed to how institutional change is facilitated or barred by the discourses policy makers and politicians use to negotiate and argue for policy reform. She distinguishes between four fundamental functions performed by discourse in the policy process. The cognitive function of discourse is to define



Case I: Sickness Insurance  

 223

the purposes of policy reform and the problems it is set to solve and offer policy instruments and methods appropriate to address this. The normative function of discourse involves demonstrating the policy programme’s consistency with political goals and ideals, for example by appealing to national values and identities. While these two functions belong to the ideational dimension of discourse where policy proposals are framed in ways that may or may not appeal to our brains and hearts, respectively, the next two functions make up the interactive dimension of policy discourse. Of particular importance to our analysis is the coordinative function, which means that a common language (epistemic community, discourse coalition) is created whereby different actors can communicate and come to agreement on a specific policy programme. Finally, the communicative function of discourse is about how policy actors are able to communicate a policy programme to the public.

12.3 Case I: Sickness Insurance The Norwegian sickness insurance scheme is often labelled “the most generous in the world”, and, currently, it probably is. It entitles employees to full wage compensation for up to a year.1 The employers pay for the first 16 days, and thereafter National Insurance carries the full financial responsibility. Employees can self-certify the first three days of absence – more if their employer is part of The Agreement for a More Inclusive Working Life (IA-avtalen). Remarkably, this scheme and the associated economic incentive structure have remained nearly2 unchanged since its introduction in 1978. The scheme was introduced by a reform in 1977. It was pushed forward by the left and the labour movement but eventually passed in Parliament by a unanimous vote. It gave all workers access to sick pay benefits at a 100 per cent replacement rate, something which many whitecollar workers had already enjoyed for years. The concerns that existed over potential abuse were trumped by arguments for equal treatment of blue- and white-collar workers. There has, in the decades since the scheme was enacted, been no shortage of demands for and attempts at retrenchment reform. At several points in time, especially when sickness absence figures have been on the rise, public commissions have been appointed to address the issue. They have pointed to problematic aspects of an economic incentive structure that, firstly, gives few incentives for employees to return

1 The insurance is paid out up to a maximum ceiling of six basic amounts, currently about 500 000 NOK/66 000 EURO. Many employers, including the state, also top this up so that a large share of more high earning employees also enjoys a 100 per cent replacement rate. 2 The employer period was extended from 14 to 16 days in 1998. There have also been some minor adjustments of entry criteria, maximum ceilings and so forth, but the main principles of 100 per cent sick pay and two weeks of employer funding remain intact.

224 

 To Reform or Not to Reform?

to work, and, secondly, does not induce employers to take steps to make workplace adaptations and other measures to bring sick employees back into work (NOU, 1990, 2000; Mykletunutvalget, 2010). Both left-wing and right-wing governments have, typically in association with the preparation of the annual state budget, made formal proposals to extend the employer financing period and, more rarely, to introduce an element of self-risk for employees. But resistance has been fierce, and all claims for change have ultimately failed. In the case of employer financing, a significant argument has been the fear that higher costs will make employers more reluctant to employ persons with poor health, thus excluding persons with impaired health from the labour market. With respect to sick pay cuts, it is a powerful argument that these will hit women and low-paid, low-educated employees the hardest, as these are the groups that tend to have the highest levels of sickness absence. In any case, Norwegian sickness insurance seems to provide a strong example of institutional inertia. Instead of benefit cuts and changes in the incentive structure facing employers, a policy strategy has been pursued emphasizing tripartite collaboration (between the state and the social partners) for more efficient workplace-orientated measures against sickness absence. In 2001 conflicts over possible sick pay cuts were resolved through the establishment of the first IA-agreement. The state and the largest employers’ and employees’ confederations joined forces to reduce sickness absence. Workplace-orientated measures emphasizing early intervention, monitoring and dialogue between employers and employees were pursued in combination with a move away from diagnosis towards functional assessments and functional ability – in sum encouraging an approach where adaptations and cooperation in the workplace should enable people to stay at work and return to work sooner despite health problems (Hagelund 2014). The agreement rested, however, on the condition that the government would not make any further claims to change the economic incentives in the sickness insurance. Thus the social partners had effectively been set up as veto-players through tripartite negotiations, with limited public debate. Despite its failure in reaching the original target of a 20 per cent reduction in sickness absence, the IA-agreement has since been renewed twice. Furthermore, measures established in the context of IA-negotiations have also been incorporated in general law, thus applying to the entire labour market. These include increasingly detailed regulations of the steps to be taken to follow up sick listed employees through individual plans and compulsory meetings. In this sense institutional change is taking place but in an incremental manner which neither takes the shape of radical reform or affects workers’ rights to sick pay. Instead layers of new rules and regulations are added on top of the existing sickness insurance. Indeed, it seems that it is this very process of layering that has enabled the state and the social partners to establish a new consensus about protecting sickness insurance from retrenchment.



Case II: Pension Reform 

 225

12.4 Case II: Pension Reform The existing Norwegian old-age pension system is less spectacular from a comparative perspective. The main features of the old “National Insurance” system date back to the late 1960s when a two-tier system of public pension provision was introduced combining an existing universal flat-rate benefit with an ambitious new earnings-related second tier. While the reform was eventually adopted with broad parliamentary support, the original impetus came from organized labour. The Norwegian LO (the Norwegian confederation of trade unions) had since the late 1950s demanded improved income protection for blue-collar workers and a levelling of pension rights with white-collar workers and civil servants who were covered by voluntary, employer-financed pension schemes (Pedersen 1990; Hippe & Pedersen, 1996). The introduction of an earnings-related second tier was believed to make additional occupational pension coverage redundant, but as it turned out supplementary occupational pension schemes continued to play an important role, particularly for employees in the public sector. Concerns about the long-term fiscal sustainability of the National Insurance system had already begun to surface in the early 1980s. A process of policy deliberation initiated by a centre-right government but later taken over by a Labour government resulted in the enactment in 1989 of a smaller parametric reform that involved modest cuts to future replacement rates. As a consequence the importance of supplementary occupational pension schemes grew even more. In the 1990s opportunities for early retirement were expanded. In 1988, at the onset of an economic downturn, the LO managed to persuade the employers association (NHO) and the Labour government to establish a contractual early retirement scheme (the so-called AFP-scheme) that allowed employees covered by a collective wage agreement to retire before the normal retirement age of 67 without any actuarial penalty. In the following years the early retirement age in the AFP scheme was lowered in several steps to reach 62 years in 1997. The gradual expansion of this scheme, which heavily subsidized early retirement, was reluctantly supported by successive governments, but concerns were growing among policy analysts and the political elite – including the leadership of the Labour Party – about declining effective retirement ages. The current pension reform process started when a Labour minority government in 2001 appointed the so-called Pension Commission with a mandate to investigate the need for a major reform of the entire system of income protection in retirement. The Commission included prominent representatives from each of the parliamentary parties as well as a number of independent experts. Contrary to standard practice, the social partners were not directly represented in the Commission and were only invited to join an adjacent consultative body. The Commission published its report in early 2004 (NOU, 2004, p. 1). It contained a majority proposal for a comprehensive structural reform of the National Insurance

226 

 To Reform or Not to Reform?

pension system. In the following years the reform was carried through the political system by shifting governments with only small modifications. It was finally enacted in 2009 by a broad coalition of parties - excluding only the populist right party (the Progress Party) – and the reform took effect from 1 January 2011. The content of the reform is strongly inspired by the innovative Swedish (and Italian) pension reforms from the previous decade. In the new system the accrual of old-age pension rights will follow the Notional Defined Contribution (NDC) formula. However, despite the transition from a traditionally defined benefit to a defined contribution formula for the accrual of pension rights, the overall distributive profile of the new system is not all that different from the old system, as both the ceiling on yearly earnings and the Guarantee Pension help maintain a fairly progressive overall benefit profile (Christensen, Fredriksen, Lien & Stølen, 2012). In addition, the reform introduced a flexible retirement age between 62 and 75 years on actuarially neutral terms, which means that the new system provides very strong incentives to postpone retirement. In the national wage negotiations in 2008 the government managed to persuade the social partners in the private sector to transform the existing private AFP-scheme so that it also conforms to the principle of actuarial neutrality, and, hence, in the private sector all subsidies for early retirement have been effectively removed. The reform contains two retrenchment measures of which the first – life expectancy adjustment – is the most important. Old-age benefits will in the future be reduced in proportion to observed increases in longevity. It is estimated that the life expectancy adjustment will result in a 20 per cent reduction of pension benefits for the cohorts retiring around 2050. Thanks to the flexible retirement age, however, pensioners can in principle compensate for the reduction in yearly benefits by postponing retirement and working longer. The other retrenchment measure concerns the indexation of pension benefits that are no longer supposed to take full part in general real wage increases. It is estimated that this measure will achieve a reduction in pension expenditures of about 7 per cent compared to full wage indexation. It is projected that the reform will result in a reduction in projected pension expenditure in 2050 from 15 to 12 per cent of GDP. The Norwegian pension reform is remarkable because it has been carried through despite rather modest reform pressures. Public expenditure on old-age pensions is currently relatively low due to a combination of moderate benefit levels and a comparatively high effective retirement age, and the Norwegian state is in an extremely fortunate situation with yearly surpluses on the budget in the area of 15 per cent of GDP that are set aside in a state-owned provident fund – known as the Norwegian Petroleum fund (renamed the Government Pension Fund Global in 2006) – and invested in international capital markets. However, a major setback in the pension reform process happened in 2009 when the government failed to persuade the public sector unions to accept a reform of the public sector version of the AFP-scheme and the public sector occupational pension



Actor Constellations and Strategic Games  

 227

schemes. This means that two main principles of overall reform – transition to an NDC-formula with lifelong accrual of pension rights and a neutral system for drawing benefits – have not been implemented in the public sector, which in Norway accounts for roughly one-third of the workforce.

12.5 Actor Constellations and Strategic Games Power mobilization theory and a rational choice perspective on policy making would lead us to believe that organized labour and the Norwegian LO in particular would be strongly opposed to retrenchment reforms in both sickness benefits and old-age pensions. As we have seen, in both areas the existing welfare arrangements have come about as a result of political processes in which the LO was heavily involved – in the 1960s and the 1970s, respectively. In fact, both the existing sickness benefit system and the public pension system can been viewed as the realization of historical claims made by the LO on behalf of its constituency of wage earners. How can we explain, from a rational power-resource perspective, the fact that the LO has succeeded to insulate the sickness benefit scheme from all attempts at cut backs, while it on the other hand has been forced to accept an ambitious structural pension reform? Clearly a general weakening of organized labour as a political force, that Pierson seemed to take for granted in the original formulation of his New Politics of welfare thesis (Pierson, 1996), would not account for the differential outcome, and this diagnosis simply does not fit the Norwegian case. In the 1990s tripartite cooperation was strengthened in Norway, and successive governments depended heavily on cooperation from the LO to deliver the necessary wage restraint to help secure competitiveness and macro-economic stability (Dølvik & Stokke, 1998). The introduction of the early retirement scheme (AFP) and its expansion over the 1990s is a clear indication of the substantial bargaining power of the LO vis-à-vis the employers association (NHO) and successive governments. We therefore need to take into consideration 1) the priorities that that the LO has given to defend its preferred position on the different issues (sickness benefits and pensions) and 2) the strength of interests and the power resources that other important actors have brought into the process. It is apparent that the LO has signalled a strong sense of ownership of the sickness benefit programme and hence a strong commitment to defend it at all costs, while the LO appears to accept not to have a de facto veto-power over any changes in the public pension system. This differential approach to the two policy areas can be viewed as a rational prioritization and adaptation to the fact that the set of stakeholders and institutional context for reform in the two fields differ. In the sickness benefit case the LO deals with only two other main actors, the employers’ association (NHO) and the government, while the pension policy field involves a much larger set of actors

228 

 To Reform or Not to Reform?

and interests such as the pension industry, the women’s rights movement, the selfemployed, consumer and pensioner organizations, etc. Furthermore, all parts of the LO share an interest in maintaining generous sick pay. With respect to pensions, there are important internal divides that must be managed and contained, particularly between private and public sector workers, but also between male and female workers and younger and older cohorts. Finally, the policy alternatives for sickness insurance reform appear to be relatively simple compared to the complexity of pensions reform. In both cases however, the LO has been willing to accept compromises that promise to fend off less desirable alternatives. The absence of a sickness insurance reform can be explained as the result of a strategic game, where the three actors had different preferences concerning the ranking of four alternative policy outcomes. In 2000/2001, when the conflict over sickness insurance reform was at its height, there were four available policy alternatives: A. to maintain status quo – no changes in the economic incentive structure and continued tripartite cooperation through the IA-agreement; B. increased employer financing; C. reduced sick pay; D. a combination of increased employer financing and reduced sick pay. There is strong disagreement about the ranking of these alternatives between the three actors. The LO gives top priority to maintaining status quo but could probably also accept increased employer funding. The employers’ association (NHO) would ideally prefer one-sided cuts in sickness benefits, but it is not as strongly concerned with rolling back the sickness benefit scheme as one might assume (see Swenson, 2002). What the NHO fears the most are one-sided increases in the degree of selfinsurance required by employers, and hence the NHO actually prefers the status quo where the state takes the bill of sickness absence over a solution where the costs are put more heavily on employers. The NHO has however signalled that it is prepared to accept a combination of moderately increased employer funding and sick pay cuts. The latter also appears to have been successive governments’ preference, following recommendations made by a government appointed committee in 2000 (NOU, 2000, p. 27). So why has an alliance between the government and the employers’ association not been forged in an effort to overcome LO’s resistance? For both successive governments and for the NHO a head-on confrontation with the LO on this issue might not be worth the cost of jeopardizing the successful tripartite collaboration to maintain wage-constraint that helped to bring Norway out of the economic downturn of the early 1990s. For successive Labour Party governments this line of action would also involve high tactical risks. Welfare cutbacks made in alliance with the employers’ association against the will of the labour unions would threaten the Labour Party’s credibility on welfare issues. For the NHO a potential danger associated with a retrenchment of sickness benefit rights would be that it opens up the field for a radical increase in employer responsibilities on the financing side. Finally, one could ask why successive Labour governments have not introduced employer co-financing of sickness benefits against the will of the NHO. Such a move



Actor Constellations and Strategic Games  

 229

would presumably not be strongly against the interests of the LO and other labour organizations, but on this issue the LO has committed itself to take sides with the NHO in defence of the status quo. We have, in other words, a situation where the LO and the NHO have formed a tactical alliance to maintain the status quo and against any attempts by the state to give either employees or employers increased burdens. This protects them both from the less desirable alternatives of one-sided cuts that in the next round could be used as ammunition for further cuts in the other direction. Taken together, maintaining the existing incentive structures but incorporating this into the Inclusive Working Life framework can simply have turned out as the best alternative that was possible without breaking with the corporatist model of macroeconomic steering that has dominated Norwegian policy making for decades. In the field of pensions, the LO apparently has a weaker sense of ownership of the existing system. Part of the explanation is historical. The National Insurance system was introduced in 1967 with broad political support, and although the most important political initiative had been taken by a Labour government in 1963, the final legislation was carried through parliament by a non-socialist government that came to power in 1965. More importantly, interests and policy preferences are more complex, and more actors are involved in the pension field. Also differences in interests between public and private sector employees add to the complexity of the issue and make it difficult for the LO to follow a clear-cut strategy of defending the status quo. While the LO accepts being among a group of actors with a legitimate stake in the National Insurance scheme, its sense of ownership of the contractual AFP-scheme was and still is very strong indeed. When LO in the 1990s accepted parametric cutbacks in the public pension system, the organization gave high priority to an expansion of the possibilities for early retirement in the form of the negotiated AFP-scheme. The expansion of the AFP-scheme and an associated decline in effective retirement age during the 1990s was eventually considered a serious long-term threat to the national economy by experts and the party political elite, and in the late 1990s tensions were building up between the LO and the Labour party leadership over the AFP scheme and its effect on the labour supply of older workers. The ambition of the Labour Party leadership to find ways to reverse a threatening lowering of the effective retirement ages was probably among the most important motives for the appointment of the Pension Commission in 2001. The Labour government wanted to avoid a head-on confrontation with the LO that would have been the result of an isolated and direct attempt to roll-back the AFP-scheme. Instead it launched a more general pension reform process involving all the other political parties – a process where a curtailment of subsidized early retirement would be only one of many themes. Our interpretation is that the Labour Party leadership tried to form a policy cartel with the other mainstream political parties in an attempt to overcome reform resistance from organized labour and the LO in particular (Weaver 2006). When the report from the Pension Commission was published in 2004 the Labour Party leadership (at the time in opposition) decided to change strategy and seek a

230 

 To Reform or Not to Reform?

compromise that would induce the LO to support the reform. In the spring of 2005 the Labour Party leadership managed to receive official backing from the LO to pursue a compromise over the pension reform with the non-socialist parties. One of the concessions to the LO that was accepted by the Labour Party leadership and taken into a political agreement with the other political parties was that the AFP scheme should be maintained, although with stronger incentives to postpone retirement. In the private sector wage negotiations in the spring of 2008 it turned out that the red-green government that had come into power in the autumn of 2005 insisted on achieving a neutral system of claiming pensions and a complete removal of all subsidies for early retirement. After tough negotiations the government succeeded in achieving this despite strong scepticism among trade unions. This result cannot be seen as anything but a defeat of the primary policy preference of the LO. The breakthrough must, however, be seen in light of the fact that it came in a period with a very tight labour market where the possibilities for older workers to extend their career appeared to be extremely good, and hence many rank and file LO members saw their interests served by a more flexible and neutral system for claiming old-age benefits. In other words, the timing of the political bargaining process in relation to the business cycle was arguably of great importance and a good example of the importance of sequencing and conjunctural causation in time (Pierson, 2004). An appealing feature of the new system – in particular in periods with abundant work for older workers – was the abolition of the work-test and the associated possibility to start claiming your pension while continuing in full-time work. The reform was also accompanied with heavy side-payments to the LO in the form of a guarantee to maintain the state subsidies to a transformed (actuarially neutral) AFP-scheme and promises to soften the implementation of the life expectancy adjustment for older cohorts. The impressive success of the red-green government in the private sector negotiations in 2008 was followed by a rather spectacular defeat the following year as it failed to achieve an equivalent breakthrough in negotiations with the public sector unions (only some of which are affiliated with the LO). There are a number of candidates for explaining this differential outcome. One is that public sector unions had a stronger bargaining position due to stronger legal protection of their existing occupational pension schemes. In addition important public sector unions outside the LO have proved to be more uncompromising in their defence of the status quo. Finally, the timing was not beneficial. The advent of the financial crisis tended to make the employment prospects of older workers less bright than they had been a year earlier, and an upcoming parliamentary election in 2009 made a flat-out conflict with the public sector unions an unattractive alternative for the government. The government did however manage to implement one crucial component of the reform also with respect to employees in the public sector. The life expectancy adjustment factor will – at least in the long run – be implemented with equal force also in public sector occupational pensions.



Institutional Dynamics and Path Dependency 

 231

Finally it is possible to speculate that there is a direct causal relationship between the differential outcome in the two policy areas: The dedication of the LO to an uncompromising defence of the sickness benefit programme made it necessary for the organization to be more compromising in the area of pensions. To simultaneously defend both against the primary policy preferences of the ruling Labour government (supported by an even more reform-eager non-socialist opposition) might simply have been perceived as impossible or, at least, too politically costly for the LO-leadership. A similar argument can be made for the Labour governments involved. A reform of the pension system (against the primary preference of the LO) was deemed more important than a curtailment of the sickness benefit programme, and hence the former was chosen at the expense of the latter.

12.6 Institutional Dynamics and Path Dependency The idea of path dependency implies that once a specific policy path has been chosen, further steps will tend to move in the same direction. An external shock will often be necessary to instigate more radical changes. From this perspective the situation in 1978, when the new universal sickness insurance came into effect, constituted a formative moment with repercussions for decades to come. Relatively protected from external shocks in the sense of deep recession or mass unemployment, the system has been hard to change. What has once been given as a right, tends to be conceived of as precisely that, given rights as in not to be touched. But in view of this, how was the pension reform possible? Is the logic of path dependency played out differently for the two cases? One crucial difference between the two systems is their levels of complexity. The pension system is complicated. Most people are uncertain about the precise content of their rights. Rights differ vastly between individuals depending on their wages, work hours, length of careers and so forth, and public pensions are complemented by a plethora of collective and private agreements. Compared to this, sickness insurance is an incredibly simple and comprehensible system. You earn full rights after only a few weeks of employment, and the deal is both generous and easy to understand – full wages during sickness. There may even be an element of 100-per-cent-magic at play: It can appear more dramatic to change something from 100 per cent to 98 than from 90 to 80. On the other hand, there are elements about the sickness insurance design that one could expect would make it easier to change. As noted above, in an increasingreturns situation – Pierson’s stricter conception of path dependency – the costs of changing paths increase over time. Every step taken down the same track increases the chance of the next step going in the same direction. This can for example be due to large set-up costs or learning effects that makes change expensive. Extending the employer financing period or reducing sick pay are for most practical purposes rather

232 

 To Reform or Not to Reform?

simple changes that entail small costs and potentially large gains for the authorities. Indeed, this is one reason that several governments have suggested such changes as a means to bring the budget into balance. Also, unlike old-age pension, citizens have not built up their rights to sickness benefits over time, thus no complicated transitional arrangements would have to be made as has been the case in the pension reform process. The costs of changing paths in the case of sickness insurance are probably of a political rather than a practical or financial kind. Especially through the IA-agreements all the parties have invested heavily into building positive collaborative relations. The political costs of exiting the agreement would be more dramatic now, after 14 years, than it would have been in 2003 when the question of whether to continue the agreement was first raised. Interestingly, Pierson also makes an argument about how the increasing returns-logic works through ideological factors: “Once established, basic outlooks on politics, ranging from ideologies to understandings of particular aspects of governments or orientations toward political groups or parties, are generally tenacious. They are path dependent” (Pierson, 2000, p. 260). Over time the party political consensus over the IA-agreement has broadened, and a more profound belief in this type of measure is being expressed both from the left and the right. Paradoxically then, the layering of new measures on top of the original insurance that was instigated when attempts to change the economic incentive structures failed may have made such reform even harder to accomplish in the future. In the decades after the path-breaking reform in 1967 the pension system has been subject to incremental changes involving both drift and layering that have made the contemporary structural reform easier to accomplish. Due to a systematic underindexation of key parameters in the system, replacement rates never reached the original target, and this has in turn allowed occupational pension schemes to play an increasing role in the overall pension system. Up until the 1990s the LO remained faithful to the ideal behind the 1967 reform of letting the pension needs of wage earners be catered to by the National Insurance scheme, and the organization looked upon occupational pension schemes with deep scepticism (Hippe & Pedersen, 1996). After the parametric retrenchment reform in the early 1990s it became apparent that occupational pension schemes were indeed needed to achieve acceptable replacement rates for wage earners with average wages, and LO unions in the private sector started to lobby for preferential tax treatment of defined contribution occupational pension schemes as a means to achieve a higher coverage among private sector workers (Pedersen, 2000). Unions in the public sector (inside and outside the LO) were opposed to the granting of tax-privileges to (employer friendly) defined contribution schemes because they feared that this could in the long run endanger the traditional, very generously defined contribution schemes for public sector employees. When the contemporary pension reform process started in 2001, the LO was struggling with deep conflicts of interests primarily between public and private sector



Discourse and Framing  

 233

unions, and there can be no doubt that this has made it more difficult for the LO to develop and carry out a consistent strategy to defend the existing pension system against retrenchment reform. This divide between the private and the public sector became evident in 2008 and 2009 when the follow-up negotiations to adjust the AFPscheme and occupational pension schemes resulted in radically different outcomes in the private and the public sectors. In the sickness benefit case, there is no similar divide between the public and private sectors for the simple reason that the 1978 reform was so radical as to give employees in both sectors the same right to full compensation from the first day of absence. The potential dividing effect of differences in supplementary occupational benefit systems in the two sectors was effectively removed. This is an example of increasing returns logic which also affects the actors’ power resources since a higher level of generosity in sickness benefits compared to pensions creates institutional conditions that facilitate the formation of a united employee front against eventual benefit cutbacks.

12.7 Discourse and Framing Discourse and framing perspectives on policy reform stress the significance of how the need for reform is formulated and communicated in order to enable actual reform. Considering the many attempts that have been made, especially to increase employer funding, it is tempting to conclude that the need to reform has not been communicated very convincingly in the case of sickness insurance, while the need for pension reform seems to have been formulated in a more compelling manner. Vivien Schmidt (2002) suggests that such reform proposals need to attend to both the cognitive and normative aspects of discourse. Cognitive functions refer to the discourse’s efficiency in demonstrating that the proposed policy will appropriately solve relevant problems, while normative functions include demonstrating the proposal’s consistency with established values. When Norwegian governments have made proposals to reform the sickness insurance, this has often happened in the context of budgetary negotiations. Thus the need for reform has tended to be framed in terms of cutting costs and securing budgetary balance. One hypothesis may be that this constitutes a narrow cognitive type of communication, which fails to appeal to core values. One infamous example stems from 2006 when the centre-left government, represented by the Minister of Labour Bjarne Håkon Hanssen, proposed to extend the level of employer funding by making employers partially responsible for the financing of sick pay also after the first 16 days. The proposal was a result of budgetary negotiations between the coalition partners, and parts of the argumentation were about freeing funds for other good purposes: “The government has ambitions to increase funding for culture, hospitals, to young families and to road maintenance. But the practical reality for government

234 

 To Reform or Not to Reform?

and parliament is that the bill that needs to be picked up first is the costs of a rising sickness absence” (Hanssen, 2006). The attempt failed – rather spectacularly – when the employees’ and employers’ associations joined forces to fight what was perceived as a breach of the IA-agreement. It was quite noteworthy then when the Prime Minister three years later – again in the context of growing sickness absence – carefully suggested in a speech at an LO conference that new regulations of sickness absence again had to be considered: “It [sickness absence] is a serious problem. Both for the absentee, for the specific workplace which loses competence and for the community which is losing productive labour force and income and on which costs are incurred” (Stoltenberg, 2009). Here he is carefully constructing a problem which goes beyond the fiscal aspects of sickness insurance, appealing to crucial values of work and participation. The expert commission which was subsequently appointed to come up with new measures continued this kind of discourse, but its proposal of a restructured employer funding (paying less for shorter absences, more for longer full time absences) was also shelved. Thus, as proposals to extend or restructure employer funding (not to mention cuts in sick pay) tend to be dropped irrespectively of how crudely or cleverly they are framed, the ideational framing of reform proposals seems to have had limited impact. However, Schmidt (2008) also speaks of a policy discourse’s interactive dimension, which consists of a coordinative function, providing policy actors with a common language and ideational framework, and a communicative function directed at convincing the public. Through the IA-negotiations, various collaborative forums and the subsequent transfusion of IA-style measures through parliament and into the general legal framework, a very strong coordinative discourse has emerged. The discourse of workplace-oriented measures with its stress on collaboration and dialogue in the workplace, functional ability and mobilization of remaining work capacity through adaptation of work and part-time sick leave, has succeeded in providing all central policy actors with a common language that allows for consensus solutions. However, the success of this language seems to rely on the maintenance of a fragile state of equilibrium which any alteration of the basic economic incentive structures threatens to break. The Pension Commission clearly had an important function of fostering consensus among the political parties about the need for a reform that would limit the growth of pension expenditure in view of lower fertility rates and rapidly increasing longevity. The fact that the social partners did not participate in the Commission implied however that this consensus did not automatically extend to the trade unions and in particular not to their rank and file members. The Commission had published a preliminary report in 2002 that was meant to prepare the public for the eventual reform proposals (Pensjonskommisjonen, 2002). In both this preliminary report and the final report from the Commission, the economic and demographic challenges to the pension system were vividly presented (Ervik & Skogedal, 2014). Among the measures used to illustrate the need for reform were



Discourse and Framing  

 235

calculations of the dependency ratio showing that the number of economically active per old-age pensioner would decline very significantly over the coming decades. Similarly calculations were presented to show how the necessary pension contribution of wage earners (assuming that the pay-as-you go pension expenditure should be financed entirely by wage contributions) would more than double in three decades in the absence of a pension reform. Finally, the report emphasized a comparative perspective. It showed that although Norway was currently a relative low-spender on old-age pensions, this would change in the absence of reform, and in particular it was argued that Norway would be the country in the OECD with the highest growth in pension expenditure (measured relative to GDP) over the period from 2000 to 2050 (ibid.). However, the reform was not only portrayed as being necessary for financial reasons. In the reports from the Pension Commission a strong critique was raised against the existing system on justice grounds, thus also mobilizing the normative function of reform discourse. It was, for instance, pointed out how the system for accrual of pension rights, which required 40 years of contribution to receive a full pension and based the calculation of benefits on the 20 ‘best’ income years, had some rather perverse distributive effects by rewarding individuals with varying income levels over the life cycle and simultaneously punishing individuals with a more even distribution of income over the life cycle (a typical situation for manual workers). Later in the reform process and in particular in the period leading up to the crucial negotiations in 2008 about the fate of the AFP scheme in the private sector, the reform was offensively marketed as expanding freedom of choice and as welfare enhancing by allowing older workers to start drawing their pension rights without any work-test. It has been argued by Brooks and Weaver (2006), among others, that one of the political advantages of a structural over a parametric reform, and in particular of a transition to a NDC logic of pension accrual, is that it helps direct attention away from the retrenchment aspect of the reform and makes it difficult for the public to detect who are the most serious losers of the reform. In the Norwegian pension reform this aspect has been less pronounced because the decision to opt for the NDC formula was taken in the last minute before the publication of the Commission’s report and as the report offered rather detailed analyses of the distributive consequences of the reform. However, the pension reform process offers some rather outright examples of obfuscation (Pierson, 1996). The most prominent is the Labour government’s promise to safeguard the AFP-scheme while subsequently insisting that the scheme should be fundamentally transformed from an early retirement scheme to become an addition to the pension wealth of affiliated workers that can be drawn on actuarially neutral terms.

236 

 To Reform or Not to Reform?

12.8 Conclusion In this chapter we have searched for explanations for the apparent paradox posed by the successful implementation of a structural pension reform and the persistent failure to reform sickness insurance in Norway. Three theoretical perspectives from the literature on welfare state reform have been invoked, and they all appear to contribute to an explanation. Table 1 summarizes the three perspectives, each in two key points, and briefly outlines the situation for each of the two cases. The first perspective is concerned with the relevant actor constellations and the strategic games involved in each of the reform processes. A critical dimension is the part played by organized labour, in particular the LO. On the issue of sickness insurance the LO has acted with a high degree of unity and with a strong commitment to defend the status quo. Pension reform has been a more complicated issue for the organization, particularly as there are latent conflicting interests between the large public and private sector unions. The other key dimension concerns the alliances and compromises made in order to safeguard vital interests for the LO. Protection of sickness insurance has been ensured through an alliance with the employers’ confederation, the NHO. With the more complex issue of pension reform, the LO had to prioritize its resources into safeguarding certain elements, in particular the AFP-scheme, an objective that was pursued by forming an alliance with the Labour Party where the LO had to accept other key elements of the pension reform. In the end, however, also regarding the highly prioritized goal of preserving the AFP-scheme, the LO had to concede to a solution, at least in the private sector, that was strongly at odds with its primary preference for maintaining a subsidized early retirement option from age 62. What it got instead was a preservation of the AFP-scheme in name and in terms of public financial support but a scheme that now fully complies with the principle of actuarial neutrality. This is a clear-cut example of institutional change through conversion, but as a discrete rather than a gradual process (Streeck & Thelen, 2005). Finally, it cannot be ruled out that the different positions taken by the LO in the two policy domains are directly interlinked. The battles over sickness insurance and pension reform took place simultaneously, and the LO leadership might well have decided that the political costs of going strongly against the red-green government on both issues would have been too large. The second perspective was institutional theory, where we examined the significance of veto points and path dependencies. Again, in both reform processes the social partners in practice control vital veto points. In sickness insurance the establishment of such a veto point (in the shape of the IA-agreement) has been part of the policy compromise agreed upon. As such, the political repercussions of the IA-agreement from 2001 have been strong. A process of layering has taken place that in practice may make reform even harder to achieve in the future. In pension policy de facto veto points where the social partners effectively may neutralize at least parts of the pension reform exist – but only through the negotiations over occupational

(X)

X

To some extent when negotiated rights are at stake. But these are multiple and decentralized, and thus potentially working in different directions.

Reason to expect reform

X

The system is complicated to reform; Need for complex transitional arrangements; Also complicated to adjust adjacent institutional structures (occupational pensions, disability pension) in a consistent manner

Fairly easy system to The IA-agreement is reform in practical a de facto veto point terms; (But political for future reform. costs may be high in terms of breaking down trust and relations established through the tripartite IA-agreement)

X

Reason to expect reform

The LO compromises with the Labour Party in an attempt to safeguard the AFPscheme

Sickness insurance A unified and The LO and the NHO reform dedicated LO against form an alliance to reform uphold the status quo.

Divided interests within the LO

Pension reform

Path dependencies

Degree of unity and The emergence of dedication among the alliances labour unions

Veto points

Institutional perspectives

Power resource perspectives

Table 12.1: A summary of the key variables in a multi-perspective analysis of two policy reform cases

Initially not very convincing, but more elaborate problem formulations have emerged; Still difficult to obfuscate the element of retrenchment

X

Convincing problem formulation; Complexity which allows for obfuscation of retrenchment

Efficient framing

Yes, but only within the framework of the IA-agreement. Thus effectively a coordinative discourse for nonreform

Between the political parties, but excluding the social partners

Coordinative discourse

Discursive perspectives

Conclusion   237

238 

 To Reform or Not to Reform?

pension schemes. The government’s failure in persuading the public sector unions to accept reform of the occupational schemes means that central aspects of the pension reform are not implemented in the public sector. On the other hand, as these veto points are multiple and decentralized, it is also possible to play the various elements against each other, as when some public sector unions start to question whether they are also losing out on some of the up-sides of the pension reform – that is, the possibility to counter the effect of the life expectancy adjustment by working longer. The problem for the public sector unions is that they have forgone the opportunity to trade in their veto power over aspects of the occupational pension system for concessions concerning the content of the general pension reform. It is not unlikely that the public sector unions will eventually be forced back to the negotiating table as the life expectancy adjustment starts to take its toll, and when that happens their bargaining power is likely to be low. Finally we looked at how the reforms have been communicated. The pension reform was presented rather convincingly – first with reference to the future sustainability of the welfare state and later as a fairness enhancing and even welfare enhancing reform. In contrast, many attempts at reforming sickness insurance have been made more hastily in the budget process with the simple purpose of saving money. Over time, however, more elaborate formulations of the need for reform have evolved, but so far with little impact in terms of policy reform. Pension policy has also offered more ample room for creative framing and persuasion (and obfuscation) compared to the sickness benefit policy where retrenchment is more difficult to hide. The other aspect of the discourse perspective is whether an efficient coordinative discourse – a shared language in which to find policy solutions and compromises – emerges. For sickness insurance such a strong coordinative discourse has certainly evolved around the IA-agreement, but this has been a discourse entirely focused on other types of policy change than those affecting the economic incentive structure. The communicative discourse has in reality been a discourse about the need for layering in the shape of more intense follow-up procedures and activation of sick-listed employees. The Pension Commission succeeded in bringing about a coordinative discourse among the political parties on the need for reform (in the end only the Progress Party voted against the pension reform when the final bill was passed in 2009), but this consensus building did not comprise the trade unions since the social partners were excluded from participating in the Commission. This could have been detrimental to the reform process but was mitigated by the LO’s willingness to seek compromises to safeguard its most treasured interests, such as the AFP. In the communicative discourse less emphasis has been put on persuading the general public about the necessity of retrenchment and about the justification for underlying principles like the life expectancy adjustment, while more emphasis has been put on selling the reform as being in some respects welfare enhancing and in others relatively harmless (Ervik & Skogedal, forthcoming).

Conclusion 

 239

Looking at all these dimensions in combination, the contrast between the institutional inertia of sickness insurance and the path-breaking pension reform of the 2000s – a period when Norway has been characterized by economic prosperity, low unemployment and the absence of external shocks – appears less paradoxical than at first sight. A unified and dedicated LO, the alliance between the social partners, the IA-agreement as a de facto veto point and a coordinating discourse against reform constitute solid barriers towards institutional change. While the complexity of the pension system makes it more difficult to reform than sickness insurance in purely technical terms, the complexity also works in the opposite direction by creating more contradictory actor and interest constellations, as well as allowing for a discourse where the retrenchment aspects of the reform could more easily be disguised as something else. This chapter illustrates the interconnectedness of different explanatory variables, as well as the need to study the respective explanatory variables contextually. The strength of labour in the Norwegian context is undoubtedly an important part of the explanation for the resilience of the generous sickness insurance. But this strength is closely entangled with institutional arrangements inherited from the past that gives the different segments of the labour force shared interests with respect to defending it. The more complex pension system where the public pension system is combined with sector- and company-specific occupational schemes configures the interests of organized labour in a rather different manner and thus also rendered its voice(s) less decisive. In his famous article on “The new politics of the welfare state”, Paul Pierson (1996) painted a rather pessimistic picture of the ability of democratic polities to adjust mature welfare systems to looming demographic and economic challenges. If such reforms were ever to be implemented, it would require that responsible politicians be able to “diffuse” the responsibility (so that voters have nowhere to go with their dissatisfaction), “compensate” powerful interest groups, and “obfuscate” the real retrenchment involved. The sick pay controversies confirm Pierson’s predictions, while the Norwegian pension reform appears to refute his pessimistic diagnosis by showing clear evidence of a significant reform capacity within the Norwegian political system. However, it cannot be denied that the three mechanisms mentioned by Pierson have all been in use at different stages in the pension reform process.

240 

 References

References Brooks, S., & Weaver, K. R. (2006). Lashed to the mast? The politics of NDC pension reform. In R. Holzmann & E. Palmer (Eds.), Pension reform: issues and prospects for non-financial defined contribution (NDC) schemes, Washington, DC: The World Bank. Retrieved from http://siteresources.worldbank.org/INTPENSIONS/Resources/NDC_English.pdf Christensen, A. M., Fredriksen, D., Lien, O.C., & Stølen, N. M. (2012). Pension reform in Norway: Combining NDC and distributional goals. In R. Holzmann, E. Palmer & D. Robalino (Eds.), Nonfinancial defined contribution pension schemes in a changing pension world: progress, lessons, and implementation (pp.129-154). Washington, DC: World Bank. Retrieved from https://openknowledge.worldbank.org/handle/10986/9378 Dølvik, J. E., & Stokke, T. A. (1998). The revival of centralised concertation. In A. Ferner & R. Hyman (Eds.), Changing industrial relations in Europe. (2nd ed.). Hoboken, NJ: Wiley. Ervik, R., & Linden, T. S. (forthcoming). The Shark Jaw and the Elevator: arguing the case for the necessity, harmlessness and fairness of the Norwegian pension reform. Scandinavian Political Studies. Hagelund, A. (2014). From economic incentives to dialogic nudging – the politics of change and inertia in Norwegian sickness insurance. Journal of Social Policy, 43, 69-85. Hanssen, B. H. (2006). Smertelig for alle. VG, p.21, September 6. Hippe, J. M., & Pedersen, A. W. (1996). Labor movement, social policy, and occupational welfare in Norway. In M. Shalev (Ed.), The privatization of social policy? Occupational welfare and the welfare state in America, Scandinavia and Japan. London: Macmillan. Korpi, W. (1981). Den demokratiska klasskampen. Svensk politik i jämförende perspektiv. Stockholm: Tidens Förlag. Korpi, W. (1989). Power, politics, and state autonomy in the development of social citizenship. American Sociological Review, 54,309-328. Mahoney, J., & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change. Ambiguity, agency and power. Cambridge: Cambridge University Press. Mykletunutvalget (2010). Tiltak for reduksjon i sykefravær: Aktiviserings‐og nærværsreform. Expert commission report to the Ministry of Labour 01.02.10. NOU (1990). Sykelønnsordningen. Norges Offentlige Utredninger 1990:23. Oslo: Forvaltningstjenestene. NOU (2000). Sykefravær og uførepensjonering, Et inkluderende arbeidsliv. Norges Offentlige utredninger 2000:27. Oslo: Statens forvaltningstjeneste. NOU (2004). Modernisert folketrygd. Norges Offentlige Utredninger 2004:1. Oslo: Statens forvaltningstjeneste. Pedersen, A.W. (1990). Fagbevegelsen og Folketrygden, LOs målsetninger, strategi og innflytelse i pensjonspolitikken fra 1945 til 1966. FAFO-rapport nr. 110. Oslo: Fafo. Pedersen, A.W. (2000). The coverage with occupational pension schemes in Norway. Results from a survey of private sector companies. Fafo-notat 2000:23. Oslo: Fafo. Pensjonskommisjonen (2002). Mål, prinsipper og veivalg for pensjonssystemet. Foreløpig rapport fra Pensjonskommisjonen, 4 September 2002. Retrieved from http://www.regjeringen.no/ Upload/AID/kampanjer/pensjonsreformen/dokumenter/Rapport_pensjonskommisjonen_ forelopig2002.pdf Pierson, P. (1996). The new politics of the welfare state. World Politics, 48, 143-179. Pierson, P. (2000). Increasing returns, path dependence, and the study of politics. The American Political Science Review, 94, 251-267.

References 

 241

Pierson, P. (2004). Politics in time: History, institutions, and social analysis. Princeton: Princeton University Press. Schmidt, V. (2002). The futures of European capitalism. Oxford: Oxford University Press. Schmidt, V. (2008). Discursive institutionalism: The explanatory power of ideas and discourse. Annual Review of Political Science, 11, 303-326. Skocpol, T. (1985). Bringing the State back in: Strategies of analysis in current research. In P. B. Evans, D. Rueschemeyer & T. Skocpol (Eds.), Bringing the State back in. Cambridge: Cambridge University Press. Stoltenberg, J. (2009). Excerpt from the prime minister’s speech at the LO cartel conference on sickness absence. Retrieved from http://www.regjeringen.no/nb/dokumentarkiv/stoltenbergii/smk/taler-og-artikler/2009/utdrag-fra-statsministerens-tale-pa-lo-s.html?id=586591 Streeck, W., & Thelen, K. (2005). Introduction: Institutional change in advanced political economies. In W. Streeck & K. Thelen (Eds.), Beyond continuity: Institutional change in advanced political economies. Oxford: Oxford University Press. Swenson, P. (2002). Capitalists against markets: The making of labour markets and welfare states in the United States and Sweden. Oxford: Oxford University Press. Weaver, R. K. (2006) Government institutions, policy cartels, and policy-change. In I. Shapiro, S. Skowronek & D. Galvin (Eds.), Rethinking political institutions: The art of the state. New York: New York University Press.

Mia Vabø

13 Changing Welfare Institutions as Sites of Contestation Nordic welfare states have been labelled as “caring states” (Leira, 1994) and “social care regimes” (Sipilä, 1997) to signify extensive provision of publicly funded care offered to and utilized by citizens from all socio-economic groups. The provision of public services is characteristically channelled through local authorities. A broad range of statutory social services are provided by municipalities, such as kindergartens, schools, primary health care and various forms of residential care and home care. In fact local governments account for approximately two-thirds of all public expenditure and employ approximately two-thirds of all employees within the public sector (Baldersheim & Ståhlberg, 2002, p. 74). From the idea that services should be accommodated to local circumstances, considerable decision-making power has been devolved to the local level. Scholars regard the relations between the local and central levels of governance to be rather coordinated and harmonious (Baldersheim & Ståhlberg, 2002). For instance, organizing joint conferences has become a growing industry, and reform steps have to a large degree been conceived as joint central–local projects, characterized by substantial elements of mutual learning and replication across municipal borders (Baldersheim, 2003, p. 31). The Norwegian Association for Local Authorities (KS), which both takes on the role as employers’ representative and as consultant in administrative reforms, plays an important role in creating consensus between the central and local levels (Hanssen et al., 2012). During the 1990s, efforts to fine-tune the local service apparatus have increasingly taken inspiration from the global public sector reform agenda, New Public Management (NPM) (Blåka et al., 2012). Like in most Western countries, the NPM trend has attracted a lot of attention in Norway, partly because it appears as a radical “shock” suggesting that power relations should be drastically reshuffled. But partly also because the conception of NPM, as it is regarded to be a global trajectory of reform, has fuelled a range of comparative research and made researchers focus narrowly on particular practices across national borders. According to Wise (2002) this principal focus may actually have magnified its impact and distorted our view of public sector evolution. Almost all organizational changes are interpreted within a single reform paradigm, while reforms outside this paradigm go unnoticed (Wise 2002, p. 555; see also Baldersheim, 2003, p. 30). In this chapter, the Norwegian home care institution will serve as an illuminating case to demonstrate how ideas taken from the NPM agenda intersect with other drivers of change rooted in other normative positions. The chapter demonstrates how this particular service has been overlaid by expectations from central and local governments. Some elements in this process has been legislative in character. However, © 2015 Mia Vabø This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



The Contested Character of Welfare Service Institutions 

 243

within the multi-level Nordic democracy, there are multiple channels, which allow a wide range of actors to influence the policy process (Burau & Dahl, 2013; Meagher & Szebehely, 2013). The chapter particularly illustrates the way in which NPM-related ideas have provided points of leverage not only for actors driven by the search for efficiency, but also for actors concerned to rearticulate more traditional public sector values. Before turning to the empirical part, the chapter starts by mapping out a conceptual framework through highlighting social service institutions as contested formations, characterized by modes of governance that are not readily compatible. The framework also highlights the process of change as a discursive struggle – a process whereby various actors aim to create, maintain and disrupt institutions in accordance with their particular frames of meaning.

13.1 The Contested Character of Welfare Service Institutions In their search for conceptual models that adapt endogenous as well as exogenous sources of change, Mahoney and Thelen object to the widespread assumption that institutions are almost inert formations. In their approach, institutions are fraught with tensions because they inevitably raise resource considerations and invariably have distributional consequences. Hence, there will always be some actors who are in favour of change, while others may defend the status quo (Mahoney & Thelen 2010, p. 8). In the case of public social service institutions, it is widely recognized that ongoing and persistent tensions exist, for example, between fiscal goals and welfare goals, between efficiency and various probity measures, between individualized care and procedural justice (see for instance Lipsky, 1980; Gummer, 1990; Harmon & Mayer, 1986; Newman, 2001; Hogget, 2005, 2006). In order to highlight that a number of people are bargaining among themselves to decide on the best way to get things done, organizational scholars have characterized social service organization as multiple actors/multiple goals organizations – in sharp contrast to single actor/single goal models (Gummer, 1990, p. 9). Tensions between different ideas about the best way to run services are finemeshed. Care professionals may disagree, for instance on whether care providers should take on an enabling or supportive role, or they may disagree on how service provision can best be specialized and coordinated. Hood and Jackson (1991) argue that, even though arguments about the best way to run public administrations are numerous, these arguments seem to serve a small but stable set of administrative values. They distinguish between three families of administrative values – all of which are widely accepted, but still difficult to satisfy by the same organizing principle. The first set of values is connected with the matching of resources to defined objectives – that is, to maximize outputs. The central concern is to keep public organizations lean

244 

 Changing Welfare Institutions as Sites of Contestation

and purposeful – to trim fat and avoid slack. An orthodox organizational design for realizing these values is the rational goal model – a typical mechanistic structure and setting of fixed and checkable goals. The second set of values is related to rights and the adequacy of governmental process. The core values of fairness, equity and rectitude are ensured through process control rather than output control. Hence, the ideal organizational form is the hierarchy model characterized by transparency, due process and formal accountability flowing upwards to democratic bodies. The third set of values includes traits such as resilience, adaptivity and reliability. Public service providers are expected to be able to operate even in worst-case conditions, to adapt rapidly in a crisis and to be able to learn from failures. Adapting to ambiguous and unstable conditions calls for an organic organization – based on a high degree of slack, committed staff, cooperation and constant adjustments. This third set of values has been regarded as a more “hidden” constitutional dimension of government associated with public trust (Toonen, 2007, p. 305) and the way discretionary power of public service agencies is used to create “legitimacy from below” (Rothstein, 2007, pp. 213-221). Put briefly, it may be related to the ability of institutions to avoid scandals and negative headlines in the news. According to Hood (1991) these three sets of values are regarded as indisputable though largely in conflict within public bureaucracies. For instance, it may be argued that procedural routines are needed in order to act responsibly towards all potentially worthy cases. Nonetheless, bureaucratic procedures may simultaneously be regarded as wasteful and inefficient from the perspective of the financial controller or as excessively rigid from the perspective of front-line staff, who are required to respond to special circumstances and needs of clients. The way in which the three different sets of values are balanced is not given once and for all but is constantly negotiated between different actors and may be radically disturbed when new administrative reforms are put into practice.

13.2 Institutions as Constituted by and Changed by Discourse The idea that institutions are rather stable formations, incidentally punctured by exogenous shocks, directs the focus of attention to major changes. In contrast, scholars, such as Paul Pierson (2000) draw attention to ongoing incremental changes unfolding over time because of efforts to close the gap between an institution’s intentions and outcome. Kathleen Thelen’s approach moves beyond the discussion of major versus incremental change. Streeck and Thelen (2005) and Mahoney and Thelen (2010) refer to several mechanisms of institutional change, most of which imply that institutions may change even though they appear to show stability. Institutions may change abruptly as rules are replaced by new ones, but changes often occur more gradually, as is the case with the mechanisms of conversion, drift



Institutions as Constituted by and Changed by Discourse 

 245

and layering. Conversion occurs when institutions are redirected towards new goals, functions and purposes; drift takes place when rules are formally the same, but institutions adapt to shifts in external conditions. Layering occurs when new rules are attached to existing ones – that is, that rather than replacing existing rules, new rules tend to change the structure and status of the “old” rules (see van der Hejden, 2011 for a broad discussion). My presentation of the historical institutional context of the Norwegian home care system will demonstrate the usefulness of these concepts of incremental change. Still they do not fully capture the direction and dynamic of change. Following Schmidt (2008, 2010), I believe that the dynamic of change is better explained if more attention is drawn to the substantive ideas of political actors and why they want to alter or preserve their practices. Institutions are serving both as constraining structures and enabling constructs of meaning. Institutions constitute the context within which agents think, act and speak, but they are also a result of agents’ thoughts, words and actions. Sentient agents are able to think critically about their own institutions and are able to build consensus around ideas. Hence, discursive institutionalism regards norms and rules-of-the-game as dynamic, intersubjective constructs rather than static structures. Mainstream sociological institutionalism has paid considerable attention to management ideas and the fact that these ideas are transformed as they are put into operation in different contexts. Røvik (1998), for instance, regards popular management ideas as semi-finished products and as such, required to be crafted into existing routines and practices. Discursive institutionalism takes one step further by considering the direction of change. Ideas are not free floating; they are always represented in a more overarching discourse. Any given discourse may serve to articulate not only different levels and types of ideas, but also the structure of meaning that channels policy or actions in certain directions. “A discourse refers to a set of meanings, metaphors, representations, images, stories statements and so on that in some way together produce a particular version of events” (Burr, 1995, p. 48). As noted by Schmidt (2008, p. 310), discourse is not only about what you say, but also to whom you say it, as well as how and why. It can be seen as a historically contingent body of regularized practices of language that are condoned by a particular community (De Cock, 1998, p. 2). They play an important role in structuring power relations; institutionalized practices like medical counselling or accountancy are characterized by specialized arguments and tacit rules of communication dictating what can be talked about in which ways, and (above all) who has the right to speak. Like institutions themselves discourses are neither monolithic nor static, but are overlapping and co-existing with competing discourses. Words and phrases are sometimes exchanged across discourses, filled with different meanings and used for different purposes. Words that are open to different ascriptions of meaning, so called “floating signifiers” (Laclau, 1990), tend to flourish in management reforms

246 

 Changing Welfare Institutions as Sites of Contestation

(Czarniawska-Joerges, 1993). Hence, these reforms open up for ongoing struggles between different discourses to fix the meaning of important words – for instance, the meaning of “quality” or “user orientation” (Clarke & Newman, 1993, p. 60; Vabø, 2006). In their studies of NPM-related changes in the UK, Clarke and Newman (1997) found that managerial roles and routines were layered on top of existing bureaucratic-professional practices and produced fields of tension rather than being simply replaced by new policy ideas and discourses. Newman (2001) emphasized the same points in her study of Tony Blair’s Third Way policy: “ New discourses had to negotiate or displace the residues of those installed under social democratic and neo-liberal governments, now deeply embedded in institutional norms, entrenched interests, cultural values and organizational or professional identities” (Newman, 2001, p. 166). According to Newman tensions between the “new” and the “old” may be resolved in various ways. New discourses may co-opt existing discourses into their own legitimating narrative, they may subordinate existing discourses through state power or they may be appropriated by managers, professional staff and others in their pursuit of other agendas. In the context of social service institutions it is of vital importance to recognize the distinction between changes in political and administrative decision-making and changes at a practical level of service provision. Governments may proclaim that they “have” purchaser-provider splits, performance-related pay or Total Quality Management. Official announcements, however, rarely offer a sufficient account of what is going on “on the ground” (Pollitt, 2002). Change agents – whether they are civil servants or consultants – support their arguments with metaphors, broad abstracted labels and conventional notions which permit a flexible redefinition of steps to be taken. Vabø (2002) characterizes the managerial discourse as a language of reconciliation in which seemingly opposed values are miraculously resolved. As they open up for many possible meanings, broad abstract labels become floating signifiers that may create a context of positive expectation – such as, Who can be against quality? Public management reform is often a multi-step, long drawn out and erratic process (Pollitt, 2002). An important part of the complexity is the multi-layered nature of administrative action. Professionals in social services organizations may, for instance, be offered new positions as “purchasers”, “providers” or “strategic managers” or may be subjected to new practices through inspections and audit regimes. It is, however, important to bear in mind that they are caught up in the play of competing discourses and may therefore respond to suggested changes in ways not anticipated (see, for instance, Halford & Leonard, 1999, p. 102, Harris & Unwin, 2009). People may comply with suggested changes because there is no alternative but may not always be committed to them with enthusiasm or identification. As noted by Clarke and Newman (1997, p. 54) they may be subjected to a discourse but not necessarily subjected by it. The latter way of complying with changes typically



The Institutional History of Norwegian Home Care 

 247

occurs when care staff feed data into a performance management system and thereby empower strategic managers to rationalize service provision (see below).

13.3 The Institutional History of Norwegian Home Care 13.3.1 1960s and 70s – An Era of Expansion Like many other social services in Norway, home care services originated in the third sector and were absorbed into the nationally mandated programmes after the Second World War. The service expanded greatly in the 1960s and 70s thanks to generous earmarked grants from the central government to local authorities through a system whereby considerable costs for care were reimbursed. However, as an infrastructure of services was established by the end of the 1970s, the active role of the state and incentives for expansion were replaced by incentives for finding cohesive and cost effective care solutions at the local level.

13.3.2 1985-1995 – An Era of Decentralization and Democratization In the mid-1980s the central government delegated to municipalities the responsibility for a whole range of services. The generous reimbursement system was replaced by one based on block grants. These efforts of devolving functions downwards went hand in hand with a reform strategy stressing participation and local democracy. A core argument behind the reform was to integrate citizens in the governing of public services in order to make services adapted to local conditions. Although it was never questioned whether home care services should be offered as an entitlement of citizenship, national policy documents valued partnership working with families and active co-responsible citizens (see also Sehested, 2002; Vabø, 2011b) As subsidies from central government were capped, local authorities had to curb expenditures. Most local authorities then reduced the number of beds in institutions and transferred typical nursing tasks from institutions to the home care system. This marked the beginning of a process of change whereby the role of home care changed from a preventative role stressing practical and social care for the elderly with moderate care needs towards a more medicalized role of providing rehabilitation, medical assistance and nursing care for the frail and sick elderly and terminally ill. Seen through the lens of Mahoney and Thelen’s concepts these changes may be characterized partly as a process of conversion – a redeployment of old institutions to new purposes. Changes have clearly also occurred through institutional drift as domestic and social tasks have tacitly been off-loaded from the home care system without any political decision-making or legislative changes (Vabø & Szebehely, 2012).

248 

 Changing Welfare Institutions as Sites of Contestation

The decentralization reform also encouraged local governments to organize services in a cost-effective manner. Home care was typically organized in selfregulated teams, providing services for the population of a specific geographical area. Organizational arrangements reflected a belief that human service organizations are bound to deal with needs that are contextual, complex and shifting. It was argued that those closest to the individual service recipients (Olsen, 1992, p. 124) therefore should conduct needs assessments and service allocation. The team organization corresponds with Minzberg’s (1983, p. 105) notion of horizontal decentralization – the transfer of responsibility from line managers into the realm of operators – that is, the frontline staff who work directly with service recipients. Viewed against Hood’s classification of administrative values, the arguments behind the team organization were linked to core values of resilience and reliability.

13.3.3 1995: The Quest for Transparency and Enforceable Rights for Citizens In the years following the decentralization reform, quality problems in the local care sector became a topic of public dispute. Public debates on eldercare peaked in 1990 during the so-called “elderly revolt” – a media protest started by a middle class senior citizen (Vabø, 2011b). The heated public debates generalized a kind of media hype portraying individuals’ grievances against local services, often accompanied by comments from various actors entering the role as champions of the elderly in need of care. Unlike the ideal of active co-responsible citizens figuring in policy documents associated with the decentralization reform, the citizen of these media hypes was always a possessor of rights – a taxpayer with the right to expect top-quality service from a wealthy oil-nation. In the 1990s, the legal security of citizens and accountability of local governments was put on the agenda by central governments. A range of measures were implemented to enhance the transparency of local service provision and to make the legal rights of citizens more enforceable. Procedural rights of citizens were strengthened as the Social Services Act stipulated the right of citizens to have their needs individually assessed, to make their views known, to receive written and wellfounded decisions and the right to appeal. Furthermore, a new clause in the Health Services Supervision Act of 1984 was added in 1992 requiring that health and care services should implement a system of internal control (internal audit). The focus on internal control systems was endorsed by a Quality Regulation (kvalitetsforskriften) directed towards all agencies providing long-term care (Ministry of Health and Social Care, 1997). All these regulations called for more explicit allocation criteria and work procedures in home care. Local care agencies were encouraged to put into writing what was previously carried around in the heads of front line staff (Vabø, 2011a). Local governments were also increasingly required to report information about finances, governance and services. A national information system, KOSTRA, was put



The Role of NPM as a Driver of Change 

 249

into practice. The system was based on consecutive data records and annual reports from local authorities and made it possible to compare resource use and broad quality indicators across municipal borders. An additional reporting system, IPLOS, was later added to the system, providing encrypted individual information about all recipients of care in Norwegian municipalities. The quest for transparency and improved control was clearly rooted in traditional public sector values, stressing rights and the adequacy of governmental processes. However, central health authorities also encouraged municipalities to utilize data from KOSTRA-IPLOS for benchmarking and learning from “best practice” municipalities. It became highly entangled with a modernization agenda inspired by NPM.

13.4 The Role of NPM as a Driver of Change Overlapping with the increased control from central governments in the 1990s, a range of measures associated with the global NPM agenda became a part of a general strategy of public sector modernization in Norway. A report commissioned by a right-wing government (NOU 2000) stated that competition is the key to quality. Accordingly, various forms of competition were recommended for public services such as competitive tendering, free-choice models and benchmarking (NOU 2000, p. 19). True marketization – that is, the idea to contract out services to for-profit companies – became a matter of passionate public dispute; in particular within elder care (Vabø et al., 2013). The resistance against “putting grandmother out to tender” was fronted by a large and well-organized trade union movement, The Norwegian Union of Municipal and General Employees (“Fagforbundet”). The union worked (and still works) systematically with awareness around marketization issues – working from below with trade union representatives rather than lobbying in Parliament. In addition, the unions felt the need to come up with constructive alternatives because they saw that if they were simply seen as arguing against change, they would have little impact. The most prominent result of the search for alternative development strategies is the Model Municipality Experiment, a counter strategy based on a basic view that there is a close link between working conditions and conditions for good services. The idea was that skills, knowledge and initiatives from “below” could contribute to enhancing cost-efficiency and quality – cooperation and not competition would produce better, more cost-effective services. The approach was based in the previous policy discourse stressing empowerment of staff and citizens (Vabø et al., 2013). This counter strategy succeeded in prevented the marketization discourse from fixing the meaning of quality. Even though “true marketization” was contested, a range of softer NPM measures have been adopted, at least by urban and densely populated municipalities. Efforts to modernize the public services have been influenced by ideas such as purchaserprovider splits, autonomous budgetary units, fee-for-service reimbursement and

250 

 Changing Welfare Institutions as Sites of Contestation

various forms of marketing and quality management systems (Torsteinsen, 2012; Blåka et al., 2012). A core model associated with the NPM inspiration in home care is the purchaserprovider model – a model suggesting that responsibility for assessing and approving a contract for services should be separated from the responsibility of providing services (Vabø, 2006). The distinction between purchaser and provider roles is regarded as a necessary step towards competitive tendering and the purchase of services from private agencies. However, the Norwegian Association of Local and Regional Authorities (KS) argued that the separation and specialization of roles would also be appropriate for those who did not tender out services. KS argued that it would make in-house providers better positioned to demand quality and to control and manage quality at arm’s length (Pape, 2000). What is interesting to note here is that the purchaser-provider model was also justified as a way for local governments to deal with the legal and formal aspects of service provision. Many local authorities were already pressed to find adequate ways to adapt to the procedural rights of citizens stipulated in the Social Service Act (see further details below). Hence, in contrast to the NPM textbooks’ focus on the model as “managing by contracts” (output) rather than by rules (input), local administrators argued that the model was an appropriate structure for dealing with new procedural rules decreed by law (Vabø, 2011a). They believed that specialized care assessors would be apt to take a more detached view of care needs than would the hands-on care staff (Gammelsæther, 2006; Vabø, 2007). Hence, the purchaser-provider model was reframed within the discourse of the classic rule-bound bureaucracy. This reframing obviously added legitimacy to the model and contributed to making it more acceptable and more widespread (see Blomberg, 2004 about similar trends in Sweden). The purchaser-provider model became an assumed cure-all model – a model able to enhance cost effectiveness as well as procedural justice.

13.5 Processes of Change at the Operational Level The brief historical account suggests that the Norwegian home care institution has been characterized by a process of layering, whereby by new modes of governance stressing vertical integration – both in terms of more regulated routines for service allocation and in terms of contractual control from “above” – have been overlaid onto an organizational form characterized by horizontal decentralization. Questions then arise concerning how these new ideas of governance intersect with the norms and values justifying the “old” team organization. How did hands-on care staff understand and respond to the new imperatives suggesting that their relationship with clients should be guided by formal agreements and standards specified by specialized needs assessors?



Self-regulated Care Teams Facing the Purchaser–Provider Model 

 251

I now turn to the practical level of the home care institutions and to the lived experiences of managers and care staff who have been the target of reform. This section draws on findings from my own published case studies conducted in several urban home care districts over the past 20 years. The first case study, conducted in the mid-1990s – the “Care Team Study”– provides insight into the team organization model which was the typical way of organizing home care the first decade after the decentralization reform (Vabø, 1998). The second study – the “Reorganization Study”– sheds light on how the efforts to restructure the home care service in line with softer NPM ideas was perceived and responded to by managers and staff. More recent studies, the “Follow-up Studies”, were conducted between 2007 and 2010 and intentionally selected to illuminate how statutory accountability requirements and NPM-related measures have interacted and affected routines and service allocation. (For further details see Vabø 2011a; 2012.)

13.6 Self-regulated Care Teams Facing the Purchaser–Provider Model My inquiries into five different home care teams in the mid-1990s confirmed that care teams were based on a collegial organization characterized by vertical communication and practitioners and staff managers (group leaders) working cooperatively. Decisionmaking power was delegated, not only to skilled nurses and auxiliary nurses, but also to lower-skilled home helpers who spent more time with clients and thereby acquired first-hand knowledge of the everyday life of clients. The group leaders normally made the initial contact and oral agreement with new clients either in a hospital or in the client’s home. However, the initial agreement often had to be adjusted or altered because care needs were ambiguous or unstable at first. Care staff interviewees argued that at the very moment elderly people are discharged from hospital, they may need time to recover and time to experience what their body is able to manage. Care needs may also suddenly shift due to changing health conditions or changing circumstances in the standing arrangements of the family. Both group leaders and managers regarded it a matter of course that needs assessment was a continuous process overlapping with service provision. Senior care staff who had worked in home care since the 1970s regretted that available resources had become more scant over the past years, and as a result, service staff had become preoccupied with allocating these scarce resources in a manner that would ensure that the most urgent needs were prioritized. Their core working principle was to enable people to utilize their own coping capacity. Hence, people processing work was very much about identifying needs (“real needs”, “urgent needs”, “no needs”) and classifying people according to their willingness and ability to utilize their own coping resources.

252 

 Changing Welfare Institutions as Sites of Contestation

As their work had to be carried out within the limits of scarce resources, the degree of attention devoted to individual care recipients was regularly diminished by the needs of other service recipients and by the unpredictable influx of new clients requiring the caregivers’ help. In order to balance competing needs, care staff constantly had to engage in negotiations and small talk with service recipients to make the daily work run smoothly. They regarded this kind of “interactional work” (Strauss et al., 1985) as a necessary ingredient in their work although they generally experienced the juggling between competing needs an unpleasant part of home care work. Interviews with elderly service recipients echoed the logic of justice expressed by care staff: The majority of interviewees expressed great sympathy for the busy working conditions of care staff and for other care recipients whom they believed had more urgent needs than themselves. However, a minority refused this frame of reference. In their opinion home care services was a self-evident right of citizens and therefore services should be carried out according to clear entitlement criteria and incontrovertible agreements. These arguments, from allegedly demanding care recipients, echoed the quest for transparency displayed in the heated media debates at that time (see above). Accordingly, the push for transparency was not just an idea from a policy agenda “from above”; it was also rooted in people’s lived experiences within the care system. Shortly after the “Care Team Study” was completed, the purchaser-provider model was for the first time introduced in the Norwegian care sector. The abstract principle of separating the responsibility for purchasing services from that of providing those services implicitly presumes that needs are fairly stable and unambiguous, at least for those equipped with professional skills and adequate assessment instruments. It also implicitly presumes that needs assessment should be based on more explicit criteria and predefined needs categories and not on moral judgments concerning the coping capacities of named individuals (Vabø, 2011a). In the “Reorganizing Study” conducted at the turn of the century (Vabø, 2002) I got a taste of the first reactions among manager and staff who had recently modernized their work organization in line with the purchaser-provider model. As mentioned above, the new structural arrangement was justified as a way to strengthen the legal rights of care applicants. Care managers and staff had been told that specialized care assessors would relieve them from paper work and make sure that citizens were well informed and equipped with a proper letter of assignment. Even though these arguments were widely accepted as legitimate, care staff still regarded it as a matter of fact that specifying needs prior to the service provision is often next to impossible. Even the needs assessors themselves, who had previously worked as group leaders of home care teams, recognized that care needs are inherently ambiguous, unstable and shifting. They experienced the new assessment routines as cumbersome. In addition to attending to the frail elderly person in their charge, they had to be attending to procedures and form filling. When confronted with weak



Self-regulated Care Teams Facing the Purchaser–Provider Model 

 253

patients who had just been discharged from hospital, they sometimes felt rude in “bombarding” their patients with questions and written information. They knew from previous care work experience that many frail elderly need some time to be aware of their post-discharge condition before they are willing to admit they need help. Care staff made ironic comments about new legal protection talk and gave narrative examples of how written information provided for legal protection sometimes gave rise to anxiety among frail elderly people because they did not have the energy to read them through. Nonetheless, they found it improper to make their voice heard. “Who can be against legal protection of vulnerable old people?” Soon after the new structural arrangement was settled, provider staff had experienced that the separation and distance between needs assessors and hands-on staff was problematic – particularly when facing sudden shifts in care needs or facing care needs that had not been fully recognized by the needs assessor. New routines required a formal reassessment procedure. If the amount of service offered was inadequate, they had to state a reason for their judgment, make a phone call to the needs assessors and request a reassessment. The new procedure was awkward, laborious and sometimes unworkable because of time constraints. Problems relating to inflexible and time consuming assessment routines cleared the way for a softer division of work between the needs assessors and care providers. Shortly after the model was implemented routines were renegotiated among local managers. Agreements were made to adopt a “wait-and-see” approach, which meant that care assignment letters and more fixed care routines were not settled prior to service provision, but after two weeks – after the opinion of care staff had been heard. Additionally, as unstable and unpredictable needs continued to occur, the terms of the contracts were regularly ignored. In certain situations it appeared inhumane and it would obviously result in blatant disregard of the wants and needs of the care recipients not to give a hand with small tasks, whether it be the laundry, feeding the wild birds outside the window or getting some extra groceries from the shop. The contractual logic of the purchaser-provider model was softened in order for care staff to attend to “real needs”, “urgent needs” and “unstable needs”. They regarded their own adjustments in the day-to-day practice as “rescue operations” when confronted with frail elderly who were dependent on help. Ironically, their rescue operation may also be seen as a way to hide unfortunate consequences of the purchaser-provider model and thereby an indirect way to sustain the prevailing discourse and direction of reform (Vabø, 2006). Or, to use Newman’s concept, their rescue operation may have served to co-opt the professional discourse and practices into the legitimizing narratives of NPM. While both purchaser and provider staff interpreted the new roles and routines in a heuristic manner, officials of the city district administration were highly committed to the aim of modernizing the service apparatus. Two change agents (called “internal quality consultants”) were working full time to flesh out the bones of the new purchaser provision structure. Their reform plans and projects were highly influenced by the NPM agenda and were to a large degree guided by the support

254 

 Changing Welfare Institutions as Sites of Contestation

services of a private consultancy company engaged by the town council. The two agents were enthused by the new world of management ideas coming from outside the municipal administration. They were involved in a process of implementing a quality management system, a regular user survey and citizen charters. They certainly realized that they had some teething troubles at the start of the project. However, according to what they had learned from private consultants as well as from a speaker from an innovative Swedish municipality, changes could take a longer time than expected, and there could be some setbacks along the way. They maintained that if more investments were made in training purchasing skills, if assessment officers were given more casework time, more training and sophisticated assessment tools, contractual control would be accomplished.

13.7 The Purchaser-Provider Organization as a Tool for Taylorization The home care district described above was also included in the Follow-up Study in 2009 – 10 years after the Reorganization study. Over time, as new work procedures and performance measures were developed, the purchaser-provider organization matured in two ways. Firstly, both staff and managers regarded collaboration across the purchaser-provider divide as inescapable. Care staff reported that they were expected to be co-responsible for making sure that reassessments were made and files updated. Officials generally expressed a more pragmatic view of the purchaserprovider split than did the devoted change agents 10 years earlier. Together with wellfunctioning computer systems, this had eased off some of the friction between needs assessors and care staff providing hands on care(Vabø, 2011a; Vabø, 2012). Secondly, services had become more strictly predefined by the purchaser, both in terms of tasks and time use. Standards and citizen charters provided information not only about entitlements; they also provided information about services that people were not entitled to expect. Hence, the new purchaser function contributed to the earlier mentioned process of drift whereby service elements were tacitly offloaded from publicly funded home care, most likely onto families and commercial service providers (Vabø & Szebehely, 2012). The trend towards standardization also contributed to change the role of professionals. Whereas the previous working principle suggested that the raison d’être of home care was to enable elderly people to promote their own self-care, home care was now more narrowly linked to the task of providing safe and sound care. Care staff appeared as responsible “doers” rather than creative “enablers” (Vabø, 2012). Furthermore, care staff had been subjected to a range of performance controls. They spent considerable time in front of computers, reporting on what they had done/not done, on how much time they had been spent on various tasks and so forth. In addition, they had to register and report IPLOS statistics (see above) and



How Responsible Purchasing Paved the Way for Professional Power 

 255

respond to various forms of audits and surveys related to quality assessments at regular intervals. The information reported by provider units was fed back up to the local administration. As strategic managers had been empowered with quantified information updating them on the performance of different service providers in the city, they constantly used data as tools to enact remedial actions. For instance, efforts had recently been made to improve the utilization of staff resources through a new division of tasks, which meant that neighbourhoods were shared differently between provider groups. According to care staff the reorganization actually undermined the efficiency of day-to-day routines as they now had to start all over again building new trust relations with new clients. The story may illuminate how the existing professional discourse, stressing flexible adaption to the coping capacity of individuals, was subordinated to a rationalistic discourse that stressed how predefined services should be delivered. Care staff adopted the new behaviours of the performance management system, but retain their self-image, values and knowledge rooted in the professional discourse. Hence, rather than melting into a new hybrid, the purchaser-provider model created tensions.

13.8 How Responsible Purchasing Paved the Way for Professional Power In general, my studies have revealed that the efforts of local governments to finetune their purchaser-provider model have not been linear (Vabø, 2006, 2009). The process of change has been oscillating back and forth between “tightening steps” (i.e., towards more formalized procedures and harder control) and “softening steps” taken in order to respond promptly to shifting needs. What is important to note is that, although the purchaser-provider model definitively has worked to intensify care work and disempower the voice of professionals, constant discursive negotiations about how to run this organization efficiently may open up new opportunities for professional discourses to influence the way in which the purchaser-provider organization is enacted. My last example will illuminate how this can happen. The topical case – the home care service in a medium-large Norwegian city – was chosen on purpose in the 2007 study to explore how a system based on “responsible purchasing” was put into practice. The decision to implement a purchaser-provider model was advertised by the local authority as a reform to secure the legal protection of citizens – not as a way to tender out services. In response to resistance from trade union representatives, the committee working to design the purchaser-provider organization came up with the term “responsible purchasing” – a term coined to emphasize that budget responsibility was transferred to the purchaser unit. This condition called for a fee-a-for-service reimbursement system – a system where the purchaser would assess individual needs and pay according to a contract specifying both the type of service offered and the amount of care staff time provided. Taking into

256 

 Changing Welfare Institutions as Sites of Contestation

consideration the lesson from other municipalities, flexible reassessment routines were built into the model. In a case of increasing or decreasing care needs, it was decided that home care staff would have to make a “Need Change Report” (NC report) to the purchaser. The care assessor would then reassess care needs and possibly change the initial contract. Care staff gradually learnt how to manoeuvre within the new system and became aware how making visible their own time use would secure remuneration that was more adequate. Staff interviewees reported unanimously that compared to the “old” team organization the new purchaser-provider arrangement caused a great deal of extra computer work mainly due to requirements to make NC reports for every minor change in care needs. For example, if the doctor prescribes for the patient to take eye drops three times a day instead of two times a day, a NC report is needed; if the prescription goes back to a two-times a day programme, another NC report is needed and so on. Even though the number of needs assessors had been doubled since the new organization model was introduced, it was impossible for them to respond to all the NC reports. They simply did not have the capacity to make all the reassessments required (see also Blomberg, 2008). Care staff, who were pressed to respond promptly to shifting needs, had to be persistent and had to spend time phoning the purchaser officers – just to get things moving. In practice this meant that reassessments to a large degree were delegated to care staff. Hence, the division of responsibility between purchaser and provider was blurred. Care staff interviewees described the system as cumbersome and time consuming. Nevertheless, the fee-for-service system relieved them of the unpleasant burden of juggling between competing needs. They did not want to return to the “old” system. They still had a memory of the unpleasant part of home care work in the previous team organization where they continually had to give less priority to some people in order to meet the most urgent needs of others. Within the new purchaser-provider organization, when new and urgent care needs occurred, they could write NC reports and allocate more resources to meet the new care needs. Hence, the new organizational arrangement had in their view solved a problem of the team organization. However, the flexibility built into the responsible purchaser-provider model certainly created a new problem for executives in the local administration as they now perceived that cost control had become very difficult. A consultancy firm, hired to investigate why costs in home care had risen sharply, pointed to several possible reasons for the problem, among which the excessive use of NC reports was one. As the consultancy firm is one of the most eager promoters of the purchaser-provider split, no critical comments were made concerning the adequacy of the purchaser-provider model in home care. The consultancy report attributed the problem of blurred lines of responsibility in terms of the skills and behaviour of employees – both of assessment officers and care staff. Officers in the purchaser unit had allegedly not managed to take the objective stance needed for their position. Their relationships with care staff had become too cosy. On their hand, care staff were accused of lacking “purchasing skills”



The Dynamic of Change in Social Service Institutions 

 257

and of having a one-track mindset on “care rationality” without any perspectives on cost control. The consultants recommended the local authority to make an effort to improve purchaser control and to restrict the number of NC reports. In particular, care providers should be made more accountable in terms of economic issues. In the wake of the consultancy investigation several suggestions were made on how to reduce the number of NC reports. One possible starting point was to make quantitative comparisons between provider units to explore whether some units were more inclined than others to write NC reports. The call for performance control indicates that the flexibility and trust characteristic of the “responsible-purchasing” model was problematized and would most likely be constrained by new control efforts in the near future.

13.9 The Dynamic of Change in Social Service Institutions In this chapter, attention has been paid to processes of change within the Norwegian home care institution. Being embedded in a nationally mandated programme on “ageing in place”, home care expanded greatly in the 1960s and 70s. The core aim of the service was to prevent and postpone institutionalization among older people. After a major decentralization reform in the mid-1980s, as the service became a part of an overall local health and social care apparatus, many local governments reduced the number of beds in institutions and thereby transferred new tasks to the home care service. Hence, the role of home care gradually changed, from a preventative role stressing practical and social care for elderly with moderate care needs towards a more medicalized role providing personal care and nursing care around-the-clock for the most frail, disabled and chronically ill (both old and young) (Daatland, 1998). Through the lens of Thelen’s concepts of incremental change, change in the service profile of home care may partly be seen as a process of conversion – a “redeployment of old institutions to new purposes” (Streeck & Thelen, 2005). Changes also appear as a process of institutional drift due to pressure stemming from processes of rationalization in other parts of the health care system – for example, in the institutional care sector as well as in the hospital sector. Domestic and social tasks have tacitly been off-loaded from the home care system without any political decision-making or legislative changes (Vabø & Szebehely, 2012). In conjunction with changes in service profile, the home care institution has been marked by various administrative reform efforts – including efforts associated with the global wave of New Public Management (NPM). As mentioned at the outset, scholars often regard NPM as a major driver of change. And it is often presumed that changes occur in a regular and unidirectional manner. However, following Mahoney and Thelen (2010) the history of the Norwegian home care system displays how measures picked from the NPM reform agenda have been grafted on to several layers of existing governance arrangements rooted in competing normative frameworks. In order to

258 

 Changing Welfare Institutions as Sites of Contestation

understand more deeply the direction and dynamic of change in these processes of layering, I argued that is useful to bring in some insight from discursive institutionalism (Schmidt, 2008, 2010). Like the theory of layering, discursive institutionalism suggests that endogenous sources of change will never enter a world of tabula rasa, but a social world already infused with meaning and its own vocabulary. My empirical examples from the home care sector give a taste of how processes of change have been driven by tensions between competing discourses and logics of governance, creating both intended and unintended consequences. For instance, I have demonstrated how the idea of competitive tendering entailed a discursive struggle around the concept of quality. The idea that competition will enhance quality may have mobilized some actors to tender out services, but it also provoked and enforced an alternative modernization agenda rooted in the normative drivers of previous reforms – stressing democracy, user-participation and a bottom-up approach to quality development. Other elements of NPM, like the purchaser-provider model, were not politically contested in the same way. Rather, as the purchaser-provider model was justified as a way to strengthen the legal rights of citizens; it operated in conjunction with government efforts to respond to people’s demand for transparency and enforced citizen rights. Hence, the core structure of NPM was “sheltered” by the classical bureaucratic discourse stressing the value of due process, transparency and predictability. My inquiries into the operational level of home care demonstrate that the purchaser-provider model subjected care workers to more rules and procedures (process control) as well as more performance control. Both these modes of governance challenged the autonomous role of care professionals and their ability to respond to the shifting care needs of individual care recipients. The way in which these tensions were resolved varied across organizational borders as well as over time. Snapshots from my case studies suggest that the contractual logic of the model is sometimes implemented in a “heuristic manner”. Care professionals collaborate across the purchaser-provider divide in order to respond to the unstable and shifting needs of individuals. In some cases the co-responsibility of care professionals appeared as hidden rescue operations taking place alongside the efforts of local governments to make service provision more streamlined and standardized. However, my case studies also demonstrate how democratic forms of decision-making at the local level incidentally allow the professional discourse to fix the meaning of buzzwords surrounding restructuring processes. The story about the Norwegian home care system exemplifies how efforts to reshape this service institution in line with ideas taken from the NPM agenda have derailed, as they have been infused with arguments and ideas linked to other reform drivers. Hence, in order to create legitimacy both from “above” and from “below”, efforts to restructure service organizations tend to have unintended consequences which in turn prompt further reform efforts. This dynamic of change suggests that the home care system has become an unstable institution, characterized by a constant striving to rebalance conflicting modes of governance.



References  

 259

References Baldersheim H. (2003). Local government reform in the Nordic countries. Bringing politics back in? In N. Kersting & A. Vetter (Eds.), Reforming local government in Europe. Closing the gap between democracy and efficiency. Urban Research International, vol. 4, (pp. 29-39). Leske + Budrich, Opladen. Baldersheim, H., & Ståhlberg, K. (2002). From guided democracy to multi-level governance: Trends in central-local relations in the Nordic countries. Local Government Studies, 28(3), 74-90. Blomberg, S. (2008). The specialisation of needs-assessment in Swedish municipal care for older people: The diffusion of a new organisational model. European Journal of Social Work, 11(4), 415-429. Blåka, S., Tjerbo, T., & Zeiner, H. (2012). Kommunal organisering 2012. Redegjørelse for kommunal og regionaldepartementets organisasjonsdatabase, [On Municipal organization 2012. Descriptions based on the organization database of the Ministry of Local Government and Regional Development] NIBR Rapport 2012:21. Burau, V., & Dahl, H. M. (2013). Trajectories of change in Danish long term care policies – reproduction by adaption through top down and bottom up reforms. In C. Ranci & E. Pavolini (Eds.), Reforms in long-term care policies in europe (pp. 79-95). New York: Springer + Business Media. Burr, V. (1995). An introduction to social constructionism. London: Routledge. Clarke, J., & Newman, J. (1993). Managing to survive: dilemmas of changing organisational forms in the public sector. In R. Page & N. Deakin (Eds.), The cost of welfare (pp. 46–64). Aldershot: Avebury. Clarke, J., & Newman, J. (1997). The managerial state, power politics and ideology in the remaking of social welfare. London: Sage. Czarniawska-Joerges, B. (1993). The three dimensional organization. A constructionist view. Lund: Studentlitteratur, Chartwell Bratt. Daatland, S. O. (1997) Social protection for the elderly in Norway. Text Series, 4. NOVA, Oslo. De Cock, C. (1998). Organizational change and discourse: Hegemony, resistance and reconstructions, M@n@gement, 1(1), 1-22. Gammelsæther, P. (2006). Forvaltningsreformer i pleie og omsorgstjenesten – en kartlegging av kommuner og bydeler med bestiller-utførermodell. Stjørdal: Ressurssenter for omstilling i kommunene (RO). Gummer, B. (1990). The politics of social administration. Managing organisational politics in social agencies. Englewood Cliffs: Prentice Hall. Halford, S., & Leonard, P. (1999). New identities? Professionalism, managerialism and the construction of self. In M. Exworthy & S. Halford (Eds.), Professionals and the new managerialism in the public sector (pp. 83-102). Buckingham: Open University Press Hanssen, G. S., Saglie, J., & Smith, E. (2012). KS i det norske flernivådemokratiet. In M. Reitan, J. Saglie & E. Smith (Eds.), Det norske flernivådemokratiet (pp.311-341). Oslo: Abstract forlag. Harris J., & Unwin P. (2009). Performance management in modernised social work. In J. Harris & V. White (Eds.), Modernising social work. Critical considerations (pp. 9-31). Bristol: The Policy Press. Harmon, M. M., & Mayer, R. T. (1986). Organization theory for public administration. Burke: Chatelaine Press. Hogget, P. (2005). A Service to the public: The containment of ethical and moral conflicts by public bureaucracies. In P. Du Gay (Ed.), Values of bureaucracy (pp. 165-191). Oxford: Oxford University Press.

260 

 References

Hogget, P. (2006). Conflict, ambivalence, and the contested purpose of public organization. Human Relations, 59(2),175-194. Hood, C. (1991). A public management for all seasons. Public Administration, 69, 3-19. Hood, C., & Jackson, M. (1991). Administrative argument. Aldershot: Dartmouth Laclau, E. (1990). New Reflections on the revolution of our time. London: Verso. Leira, A. (1994). Concepts of caring: Loving, thinking and doing. Social Services Review, 68(2), 185–201. Lipsky, M. (1980). Street-level bureaucracy. The dilemmas of the individual in public services. New York: Russell Sage Foundation. Mahoney, J., & Thelen, K. (2010). Explaining institutional change. Ambiguity, agency and power. Cambridge: Cambridge University Press Meagher, G., & Szebehely, M. (2013). Long-term care in Sweden: Trends, Actors and consequences. In C. Ranci & E. Pavolini (Eds.), Reforms in long-term care policies in Europe (pp. 55-78). New York: Springer + Business Media. Minzberg, H. (1983). Structure in fives: Designing effective organisations. Englewood Cliffs: PrenticeHall. Ministry of Health and Social Care (1997) Rundskriv om kvalitet i pleie- og omsorgsstjenestene, I-13/97. Oslo. [Directive on quality of care] Newman, J. (2001). Modernizing governance. New labour, policy and society. London: Sage. NOU (2000). Bør offentlig sektor eksponeres for konkurranse? [Should the public sector be exposed to competition?] Oslo: Arbeids- og administrasjonsdepartementet. Olsen, B. (1992). Hjemmehjelpstjenesten i endring. In S. O. Daatland & P. E. Solheim. Og du skal leve lenge i landet. Dilemmaer og veivalg i eldrepolitikken (pp. 120-138). Oslo: Universitetsforlaget. Pape, S. (2000). Bestiller-utførermodellen og kontraktsstyring av tjenesteproduksjonen. En veileder fra KS, [Purchaser-provider models and contractual management] Kommunenes sentralforbund. Pollitt, C. (2002). Clarifying convergence. Striking similarities and durable differences in public management reform. Public Management Review, 4(1), 471-492. Pierson, Paul (2000). Increasing Returns, Path Dependence, and the Study of Politics. The American Political Science Review 94(2): 251-267 Rothstein, B. (2007). Political legitimacy of public administration. In G. Peters & J. Pierre. The handbook of public administration. London: Sage. Røvik, K. A. (1998): Moderne organisasjoner. Trender i organisasjonstenkningen ved tusenårsskiftet. Bergen: Fagbokforlaget Schmidt, V. (2008). Discursive institutionalism: The explanatory power of ideas and discourse. Annual Review of Political Science, 11, 303–326. Schmidt, V. (2010). Taking ideas and discourse seriously: Explaining change through discursive institutionalism as the fourth “new institutionalism”. European Political Science Review, 2(1), 1-25. Sehested, K. (2002). How new public management reforms challenge the roles of professionals. International Journal of Public Administration, 25(12), 1533–1537. Sipilä. J. (Ed.) (1997). Social care services: The key to the Scandinavian welfare model. Aldershot: Avebury Social Service Act (1991). Act 1991-12-13-81.Oslo. Strauss, A., Fagerhaugh, S., Suczeck, B., & Wiener, C. (1985): Social Organization of Medical Work. Chigago: The University of Chicago Press Streeck, W., & Thelen, K. (Eds.) (2005). Beyond Continuity. Institutional Change in Advanced Political Economies. Oxford: Oxford University Press Toonen, T. A. J. (2007). Administrative reform: Analytics. In G. Peters & J. Pierre. The handbook of public administration (pp. 301-311). London: Sage.



References  

 261

Torsteinsen, H. (2012). Bakgrunn, diagnose, inspirasjon. In Resultatkommunen. Reformer og resultater, Oslo: Universitetsforlaget. Vabø, M. (1998). Hva er nok? Om behovsfortolkninger i hjemmetjenesten. Where to draw the line. Oslo: NOVA-rapport 8/98. Vabø, M. (2002). Kvalitetsstyring og kvalitetsstrev. Omstillingsarbeid i hjemmetjenesten, NOVArapport 18/02. Vabø, M. (2006). Caring for people or caring for proxy consumers? European Societies, 8(3), 403-422. Vabø, M. (2007). Organisering for velferd. Hjemmetjenesten i en styringsideologisk brytningstid [Organizing for welfare. Home care in an era of competing modes of governance]. Oslo: Unipub. Vabø M. (2009). Home care in transition: The complex dynamic of competing drivers of change in Norway. Journal of Health Organisation and Management, 23(3), 346-359. Vabø M. (2011a). Changing governance, changing needs interpretations. Implications for universalism. International Journal of Sociology and Social Policy, 31(4/5), 197-208. Vabø M. (2011b). Active citizenship in Norwegian elderly care – from activation to consumer activism. In J. Newman & E. Tonkens (Eds.), Active citizenship and the modernisation of social welfare (pp. 87- 187). Amsterdam: Amsterdam University Press. Vabø, M. (2012). Norwegian home care in transition – heading for accountability, off-loading responsibilities. Health & Social Care in the Community, 20(3), 225-327. Vabø, M., Christensen, K., Jacobsen, F. F., & Trætteberg, H. D. (2013). Marketisation in Norwegian eldercare: preconditions, trends and resistance. In G. Meagher & M. Szebeheley (Eds.), Marketisation in Nordic eldercare: a research report on legislation, oversight, extent and consequences (pp. 163-203). Stockholm: Stockholm Studies in Social Work 30. Stockholm University. Vabø, M., & Szebehely, M. (2012). A caring state for all people? In A. Anttonen, L. Häikiö & K. Stefánsson (Eds.), Welfare state, universalism and diversity (pp.121-144). Cheltenham: Edward Elgar. Van der Hejden, J. (2011). Institutional layering: A review of the use of the concept. Politics, 31(1), 9-18. Wise, L. R. (2002). Public management reform: Competing drivers of change. Public Administration Review, 62(5), 555-568.

Anne Lise Ellingsæter

14 Making, Unmaking and Remaking: The Evolution of Nordic Cash for Childcare Schemes Publicly subsidized childcare services constitute a basic feature of the Nordic welfare states. Childcare services are considered vital in supporting mothers’ employment and thus for promoting gender equality both in the family and the labour market. However, over the past two decades or so, cash for childcare schemes conditioned on the non-use of public childcare services have been instituted. Parental “choice” with regard to childcare arrangements – between home-based care and institutional-based care – is the common aim. Some scholars consider that this form of institutional change could risk moving the focus away from gender equality and installing new inequalities along socio-economic dividing lines. Cash for childcare benefits tend to be taken up by low income groups, keeping mothers out of the labour market and their children out of childcare services. More “choice” as a result may challenge the “traditional goals” of the Nordic welfare states (Eydal & Rostgaard, 2011a; Greve, 2011). Policy reforms which are potentially undermining these welfare states’ reputed “passion for equality” may seem puzzling. Why and how these schemes have evolved is to be addressed in this chapter. In elucidating these questions, disclosing the underlying policy drivers is crucial. Family policies are usually disputed, entailing competition over what are considered core values, probably more so than in many other policy fields (Fleckenstein, 2011; Lewis, 2008). The main assumption regarding why these policy reforms have been introduced is that family policy continues to be an important field of policy struggle among political parties in the Nordic welfare states. The role played by cash for childcare schemes in these struggles is investigated here. As to how policies evolve, the assumption is that in comprehensive family policy systems like the Nordic ones, modes of institutional change are likely to be of an incremental nature. Accordingly, an analytical framework combining theories of party competition and theories of incremental institutional change is employed (Häusermann, Picot & Geering, 2013; Mahoney & Thelen, 2010; see also Ellingsæter, 2014). Selected for comparison are the cash for childcare schemes of Finland, Norway and Sweden. All have adopted national legislation on cash for care schemes for children under the age of three that do not use publicly subsidized childcare services. Denmark and Iceland lack statutory arrangements; existing cash for care arrangements are marginal in content and reach and are seldom mentioned in comparative studies of

© 2015 Anne Lise Ellingsæter This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Political Parties and the Welfare State – Changing Relations 

 263

childcare (Rantalaiho, 2010).1 For the three countries selected, policy dynamics are examined across countries and over time. Subject to comparison are the temporality of reforms, proponents/opponents/coalitions, cognitive frameworks and modes of institutional change involved.

14.1 Political Parties and the Welfare State – Changing Relations Post-industrialism has changed the relationship between party politics and the welfare state. In advanced post-industrial economies both the electorate composition and voter preferences are likely to have shifted. Häusermann and colleagues (2013) maintain that electoral constituencies no longer correspond to those of the industrial age and that the “traditional partisan politics” theory needs revision. The latter sees parties as representatives of social constituencies (mostly industrial classes) and as bearers of clear ideological stances for social-democratic or conservative welfare policies. A linear and direct relationship between the type of party in power (e.g. social democratic vs. conservative) and policy output is assumed. By contrast, it is argued, the institutional context, party system and party competition matter for the kind of welfare policies individual parties advocate. Strategies are not always programmatic, related to a broader ideology; they are sometimes particularistic – narrowly defined to attract specific groups of voters. Thus parties with the same ideological label make different choices if they act under particularistic competition. The traditional partisan politics approach, assuming that policy preferences and choices of a party depend on the ideological family it belongs to, is considered problematic because choices may be conditional on the institutional and partisan environment (Häusermann et al., 2013). Party preferences are not necessarily stable, as some of the comparative welfare state literature implicitly assumes, but can change within a relatively short period of time (Seeleib-Kaiser, 2010). The traditional approach needs to be adapted, for instance by taking into account constraints stemming from party competition. The “strong” version of party competition theories considers parties as strategic and opportunistic; policies are often used to appeal to important groups of voters (Häusermann et al., 2013). Instrumentalist political parties can best be seen as vote and office seekers rather than driven by deep ideological commitment (Fleckenstein & Lee, 2014). Parties may also adjust their position to the position of

1 In Denmark, municipalities may offer quite limited cash for care benefits instead of childcare services (Eydal & Rostgaard, 2011b). E.g. in Copenhagen, parents with children aged six months to three years can only apply for the cash benefit if they have already applied for a place in childcare services, and the benefit can only be paid for a year. The reform has very little practical influence: Only 700 parents are reported to receive the cash benefit (Rantalaiho, 2009). Iceland has not adopted national legislation, but some municipalities have introduced local cash for childcare schemes (Eydal & Rostgaard, 2011b 2011). Information is sparse, however (Rantalaiho, 2010).

264 

 Making, Unmaking and Remaking: The Evolution of Nordic Cash for Childcare Schemes

(potential) coalition partner(s). A “softer” version builds on the socially structured voter-party link but adds competitive elements (Häusermann et al., 2013). The extent to which a programmatic goal is pursued can be limited by party competition. Moreover, with regard to family policy in particular, it is contended that policy drivers of work-family policy of the “golden age” of pioneering welfare states differ from the political environment of family policy “latecomers” (Fleckenstein & Lee, 2012). It is maintained that in contrast to the “golden age”, party competition is key to understanding family policy modernization and institutional change among the “latecomers” (ibid.). But the argument of changing political environment is also valid for the study of institutional stability and change among the once pioneering Nordic welfare states. The position of social democracy is weakened, and coalitions have been the common form of government over the past decades. Party preferences may be influenced by the parties’ relative power. Historically, in multi-party systems like the Nordic countries a left–right cleavage has been important, but the urban–rural cleavage has found expression in agrarian parties (ibid.). Moreover, in Norway and Sweden, small Christian democratic parties have played a decisive role especially in the area of family policies (Korpi, 2000). However, parties today may defend different policies than 30 years ago because of changed voter preferences and more complex political coalition patterns. Although we can no longer always deduce the preferences of a political party from its “party family”, the question is how parties might still matter (Häusermann et al., 2013).

14.2 Incremental Institutional Change – Across Nations and Time Path-dependence theory, emphasizing the importance of political choices keeping policies on the same path, has dominated the study of welfare state development in recent years. Institutions have been viewed as difficult to change: They are mostly seen as stable and only periodically punctuated by external shocks (Mahoney & Thelen, 2010). Institutional approaches emphasizing steady paths have increasingly been criticized for their static and linear assumptions (e.g. Weyland, 2008). Scholars contend that in the study of comparative politics, the focus on path-dependency and institutional stickiness has side-lined electoral politics (Morgan, 2013). The “second wave” of new institutional theories questions the persistence of institutions as taken for granted and equilibrium qualities of institutions (Hall, 2010). Hall maintains that nation states generate distinctive institutional and ideological fields that survive over long periods of time, but politics should be seen as a process structured across nations and across time: Emphasis should be on how politics are structured in each country with some attentiveness to the ways in which institutional practices, cognitive frameworks and network relations operate. Policy measures are initiated, continued, changed or discontinued by a complex set of forces. The political



Incremental Institutional Change – Across Nations and Time 

 265

imagination of a particular era – shared cognitive frameworks, ideology or ideas – are a constitutive component of social action (Hall, 2010). Time is the dimension in which ideas and institutions evolve (North, 1999, cited in Pierson, 2004, p. 1). Policy reform and impact needs to be “placed in time”; sometimes political processes and policy outcomes take a long time to unfold (Pierson, 2004). Some factors may be stable over time but divergent across cases, and the durability of some policy elements may depend on the existence of others (Hall, 2010). Certain kinds of policy outcomes may be more likely in some temporal contexts than in others, and the same causal factors may have more impact in some periods than in others – that is, there are period effects (ibid.). As an example, Hall refers to social democracy characteristically being seen as a homogeneous force operating in the same way across space and time, but the impact of social democratic governance may be tempered by cognitive frameworks distinctive to some countries and eras. Among “second wave” contributions on institutional change, an important assumption is that policy institutions are fraught with struggles going on continuously within and over prevailing institutional arrangements (Mahoney & Thelen, 2010). Institutions are laden with power implications; formal political institutions are intended to distribute resources to particular kinds of actors and not to others – this animates change. Institutional stability is a function not simply of positive feedback but of active ongoing political mobilization. Institutions are vulnerable to change not just in moments of crisis but on a more ongoing basis. Actors, coalition patterns and political conflicts are important in order to understand institutional stability and change. Where institutions represent compromises based on specific coalition dynamics, they are always vulnerable to shifts (ibid.). The expected costs of noncompliance are factored into the strategic behaviour of actors. Once created, institutions often change in subtle and gradual ways (Mahoney & Thelen, 2010). Here cash for childcare reforms are examined as modal types of institutional change outlined by Mahoney and Thelen (2010, pp. 15-18). Layering and displacement are the most relevant modes to the present analysis. Layering is the introduction of new rules in addition to or in parallel with existing rules and involves amendments, revisions or additions. Displacement is the removal of existing rules and the introduction of new ones. This kind of change may be abrupt and entail radical shifts, but it may also include a slow-moving process in which new institutions compete with old ones. An important point is that displacement may occur gradually if actors are unable to make things change as quickly as they wish. The two other modes of institutional change, conversion and drift, are generally applicable to institutions where there is a high level of discretion in rule interpretation/enforcement. The national models of cash for childcare schemes studied here vary with regard to the discretion delegated to municipalities.

266 

 Making, Unmaking and Remaking: The Evolution of Nordic Cash for Childcare Schemes

14.3 Value Competition: Equality Versus Choice Cash for childcare policy frequently has been part of the power struggles between political parties/interest groups (Sipilä, Repo & Rissanen, 2010). A common denominator of European countries that have introduced cash for childcare schemes is that centre-right governments stand behind these schemes (Morgan & Zippel, 2003). The schemes have been met with considerable opposition from parties on the political left. They are criticized for being detrimental to gender equality, undermining mothers’ employment and the development of childcare services. Struggles over childcare services versus cash for childcare to a large degree have been associated with support of a gender equality model versus a more traditional family model. Cash for childcare schemes may be used for different purposes. Those given instead of childcare services, with the intention of supporting maternal/parental childcare at home or private care arrangements are the most contested (Sipilä et al., 2010). This is the type of benefit introduced in the Nordic countries. Cash for childcare schemes are an anomaly to the earner-carer model aiming at gender equality, centred on paid parental leave and publicly subsidized childcare services. Cash for care benefits are classified as male breadwinner family policy, presuming or being neutral to a traditional gender division of labour in society as well as within the family (Korpi, 2000). Such benefits involve low payment, presuming a main (male) breadwinner in the family. The schemes are often justified in terms of more “choice” for families in arranging childcare – that is, a choice between childcare services and cash transfers for taking care of your own children (e.g. Rantalaiho, 2010). “Choice” can be underpinned by a valuation of parental care as most appropriate for the youngest children. But “choice” is also associated with a liberalist anti-state paradigm setting limits for the state’s interference in the private affairs of families; families should decide for themselves. Voter support of “parental choice” also may reflect other types of rationales, for example a desire for autonomy. “Choice” may be valued as a symbol of recognizing a variety of life styles (e.g. Giddens, 1991). While “choice” is a main rationale, cash for childcare schemes may also be underpinned by the valorization of maternal/parental care, more equal distribution of state support between families and the privatization of childcare provision (Sipilä et al., 2010). They may also be motivated by efforts to retrench public expenditure as cash for care benefits are less costly than childcare services.

14.4 Current Nordic Cash for Childcare Schemes First, a brief description of the main features/contents of the current schemes has been outlined in Table 1 (see also Duvander & Cedstrand, 2012; Ellingsæter, 2012; Eydal & Rostgaard, 2011b). Finland and Norway have introduced a state-based benefit



Current Nordic Cash for Childcare Schemes  

 267

as part of the national social security system (Eydal & Rostgaard 2011b). The level of discretion in scheme enforcement is low in Finland and Norway, but municipal policy layering, for example, added benefits to the basic state-regulated amount (Finland), or benefits provided to age groups beyond state regulation (Norway) may shape the benefits provided. In Sweden municipalities have a high degree of discretion in enforcement as they determine whether or not to offer the benefit. Moreover, eligibility criteria vary; the Swedish criteria are much stricter than in the other two countries, which may impact the households’ total disposable income.2 In terms of amounts, differences are small – the basic benefit amounts to about 10 per cent of the income of an average wage earner (Bakken & Myklebø, 2010; Eydal & Rostgaard, 2011b). Various supplements are paid in Finland, however.3 In Norway and Sweden partial use of childcare services can be combined with partial benefit. Table 14.1: Current cash for childcare schemes in Finland, Norway and Sweden

National legislation Funded by Implemented by Child’s age Partial benefit/use of childcare services Benefit can be used to buy private care Universal, regardless of other income* Amount/month

Finland

Norway

Sweden

Yes State State and municipalities 2-year-olds and under No

Yes State State 1-2 year-olds* Yes

Yes State Municipalities 1-2 year-olds Yes

Yes

Yes

Yes

Yes

Yes

No

327.46 euro

NOK 3303, approx. 430 SEK 3000, euro* approx. 340 euro

Source: Adapted from Eydal & Rostgaard, 2011b. *As of 1 August 2012 only one-year-olds: NOK 5000 for 13-18 month-olds, NOK 3303 for 19-23 montholds. As of 1 August 2014 NOK 6000 for 13-23 month-olds.

2 In Sweden, the allowance is not paid to parents who receive unemployment or sickness benefits, those on paid parental leave or those receiving asylum-seeker allowances or state pensions (Prop. 2007/08:91; Duvander & Cedstrand, 2012; Eydal & Rostgaard, 2011b). 3 In Finland, the allowance also consists of a sibling supplement (63-98 euro) for any other child in the family under school age who is cared for in the same way. In addition, every sixth municipality pays special supplements (Repo, 2010). These are the heavily populated municipalities, covering more than half of Finnish pre-school children. The supplement for one child varies between 70 and 250 euro and may also be accompanied by a sibling supplement.

268 

 Making, Unmaking and Remaking: The Evolution of Nordic Cash for Childcare Schemes

The cash for care benefit is an alternative to a place in publicly subsidized childcare services after the end of the parental leave period. Thus the potential duration of the benefit is influenced by the length of the paid parental leave scheme. The Finnish paid parental leave is the shortest (about nine months/70 per cent wage replacement). The Norwegian leave is about a year (49 weeks/100 per cent replacement or 59 weeks/80 per cent), while parental leave in Sweden is the longest (13 months/80 per cent wage replacement, plus three months at a flat rate). In all three countries, children have a universal right to a place in the childcare services: in Finland since 1996 for children under seven (for children under three since 1990), in Norway since 2009 and in Sweden since 1995.4 Maximum monthly parents’ fees for childcare have also been introduced in the three countries: 233 euro in Finland, NOK 2330 (approx. 300 euro) in Norway and SEK 1260 (approx. 140 euro) in Sweden (Eydal & Rostgaard, 2011b).

14.5 Finland: Layering The male breadwinner model was not common in Finland, and in the 1960s improved economic conditions were actually expected to lead to a reduction in married women’s paid employment (Bergqvist, Kuusipalo & Styrkarsdottír, 1999). The latter was considered negative, and a “wages for mothers” was proposed. Nevertheless, mothers’ employment was increasing, which gave rise to a gender equality debate where childcare was a main issue. In the early 1970s, the debate took on major proportions (ibid). The main cleavage was between the left and the liberal parties, favouring collective childcare and a gender equality model, and the right-wing parties supporting some form of cash benefit for mothers and a traditional family model (ibid.). Coalition governments of several parties with different views governed Finland during most of the period from 1966 to 1987 (Bergqvist et  al., 1999). The Agrarian, later Centre Party dominated Finnish politics until the 1960s, after which the Social Democratic Party became a partner in coalition cabinets. In 1972 a social democratic minority government proposed to expand childcare services towards full coverage, which won a Parliament majority (Bergqvist et al., 1999). The Centre Party unsuccessfully argued that they could not approve the proposal unless a home care allowance was also introduced. The daycare legislation of 1973 aimed at universal coverage, provided by municipalities, but demand rapidly grew more than the number of places (ibid.). Child home care allowances emerged during the early 1970s when some municipalities started to pay cash for care benefits in order to limit the growing demand for childcare services (Rantalaiho 2009). Parents’ “choice” with

4 In Sweden all four and five-year-olds have a right to free part-time daycare (Bergqvist & Nyberg, 2002).



Finland: Layering  

 269

regard to childcare was rapidly absorbed into the national family policy discourse. The pressure for a home care allowance continued, especially from the Centre Party, claiming that women in the rural areas did not need childcare services but should receive a fair share of public support for families (Bergqvist et al., 1999). After a series of negotiations, in 1984 a compromise proposal from a government dominated by the Social Democratic Party and Centre Party was approved in Parliament. It entailed the introduction of a national Child Home Care Allowance (hemvårdsstöd/kotihoidontuki) in 1985. It was presented first of all as an alternative to childcare services (Rantalaiho, 2009). But the Social Democrats refused to accept the proposal unless the Centre Party committed itself to increase support for childcare services (Bergqvist et al., 1999). Centre-right advocates managed to frame the home care allowance with positive connotations, especially the value of work at home and parents’ opportunities to choose the form of care they need (Hiilamo & Kangas, 2009). Also important to the dominant Centre Party was that childcare services did not satisfy the needs of rural families, the benefit was to compensate lack of services (Lammi-Taskula & Takala, 2009; Rantalaiho, 2010). Costs have been an overarching concern, however; cash for care is much cheaper than institutional day care (Hiilamo & Kangas, 2009). As part of the reform compromise, parents’ right to municipal daycare services was introduced together with the home care allowance. Thus the reform implied policy layering; the allowance was put on top of increased support for public childcare services. Today the home care allowance has become institutionalized as a main element of the Finnish childcare policy model. Municipal policy layering, in the form of extra supplements, provides incentives especially for low-income families to rely on the allowance as an alternative to daycare services (Repo, 2010). The allowance structure encourages families to also care for their older children at home. Some scholars contend that opposition to home care allowance was never very strong (Hiilamo & Kangas, 2009; Rantalaiho, 2010). Over the past years there have been few initiatives to abolish the benefit. Yet, other scholars argue that the consensus is “relative” and the issue of whether to develop the allowance or daycare services has constantly been a point of contention between Social Democrats and the Centre Party/Conservatives (Salmi, 2006). Moreover, the recent economic crisis seems to have set in motion a discussion about the negative effects of the home care allowance (Haataja & Valaste, 2012). In 2012 a proposal from the left-centre-right majority coalition for a partial displacement – to cut the allowance for two-year olds – caused heated public debate (Salmi & Lammi-Taskula, 2013). The motivation was both cutting public transfer expenditure and raising women’s employment rate. However, the proposal was not included in the government’s economy plan for 2014-2017 (Salmi & Lammi-Taskula, 2013). The cash for care benefit was introduced at a time marked by severe labour market problems. Over time long care leaves have gained strong legitimacy (Rantalaiho, 2009). Many mothers see the allowance positively – an extension of parental leave

270 

 Making, Unmaking and Remaking: The Evolution of Nordic Cash for Childcare Schemes

and home care is considered in the best interest of the child (Repo, 2010). This is a factor reinforcing policy stability. It is regarded as a very sensitive political issue with which no politician would like to interfere; suggestions for any cuts are characterized as “political suicide” (Haataja & Valaste, 2012, p. 2). The use of the child home care allowance has been quite stable since its introduction (Repo, 2010). Since 1990 the majority of small children have been cared for at home.5 Among mothers taking the benefit, those with less education, low income levels and many children are overrepresented (Repo, 2010). Immigrant mothers apply for the allowance slightly more often than mothers born in Finland (Eydal & Rostgaard, 2011b). Almost half of all mothers receiving the home care allowance have no real choice between paid and unpaid work as they have no job to return to, and unemployment levels have been high (Haataja & Nyberg, 2006). The Finnish form of policy layering, the notion of a universal right to childcare combined with the rather long home care of children, is seen as “paradoxical” (Repo, 2010). Finland is among the countries where employment rates for mothers with children under three are markedly lower than for mothers with older children (OECD, 2010). Fifty-one per cent of two-years-olds in Finland are enrolled in childcare compared to 89 and 91 per cent in Norway and Sweden, respectively. The benefit structure also influences the usage of childcare services among older children.6 OECD (2005) is critical of the benefit structure, noting that most of the large municipalities make additional home care payments to discourage the use of local government childcare facilities, tilting the financial incentives towards mothers choosing to stay at home.

14.6 Norway: Layering, Partial Displacement, Announced Re-Layering A “housewife wage” was the topic of debate in the first decades following the Second World War, supported by the political right wing, but also by social democrats. However, the late 1970s marked the beginning of a new era of policy initiatives to support parenthood, typifying a “modern family policy” (Vollset, 2011). The main shift, from cash transfers to services, was headed by social democracy and the left. In the mid-1970s childcare services were legislated, aiming at full coverage, with support from all political parties (except for the then small populist right-wing Progress Party) (Leira, 1992). However, the centre-right’s priority of cash for care

5 In 1998, 57 per cent of children between 9 months and 3 years received the benefit, compared to 52 per cent in 2007 (Repo, 2010). 6 68-78 per cent of Finnish children aged 3-5 are in daycare, compared to 97-99 per cent of Norwegian and Swedish children in this age group.



Norway: Layering, Partial Displacement, Announced Re-Layering  

 271

benefits over the expansion of paid parental leave and childcare services was visible in policy debates in the following two decades (Ellingsæter, 2007). Cash for childcare had been a crucial family policy goal, especially for the Christian Democratic Party. Finally a cash for care reform was introduced in 1998. Parents with children one to two years old who were not using publicly subsidized childcare services would receive about 400 euros per month. The political compromises and controversies illuminate the political parties’ different stances regarding family policy models, reflecting a leftright cleavage (see Håland, 2001), but it also demonstrates the particular role of small confessional parties in shaping family policy field (see Korpi, 2000). The reform came about as a change of government took place after the 1997 election, when the social democratic government resigned and unexpectedly was replaced by a centrist minority government (The Christian Democratic Party, the Centre Party and the Liberal Party). The reform materialized as a majority compromise between the government parties and the two conservative opposition parties (the Conservative Party and the Progress Party). This particular constellation was decisive both regarding the motivation of reform objectives and the concrete measures and eligibility criteria chosen. All these parties belong to the political centre -right, traditionally supportive of one-income families. There are important nuances in the motivation for the reform, however. Better care and more family time for children was the main concern advocated by the government. In the public debate, in which the Minister of Children and Family Affairs from the Christian People’s Party played a central role, the reform was advocated very much as a “time issue”, enhancing families’ opportunity to choose more time at home with their children. The other two objectives, freedom of choice and just distribution of state subsidies for childcare, belong to a traditional conservative family policy agenda with a preference for private to public care and cash to services. However, the freedom of choice ideal creates a tension with the ideal of family-based care because the reform content is not only valuing family time, but also the non-use of public childcare. Not only parents staying at home with their children, but also those buying childcare outside the state-sponsored services are entitled to the benefit. Thus parents may receive the benefit without spending more time with their children. Here we see the imprint of concessions made to the two conservative opposition parties; a shift toward more private childcare was their aim (Knudsen & Wærness, 2001). Whether the family or the market provides the care is value neutral within the modern liberal agenda (Ellingsæter, 2003). The social democrats and the left were fierce antagonists. The gender equality concern was prominent in the left’s opposition to the reform; it was expected to lead to a decrease in mothers’ employment and lower demand and supply of childcare services. However, in Norway the cash for care benefit was also layered on top of existing policies. Polarized views among the political parties have persisted. In the red-green majority government, in office from 2005-2013, both the Social Democratic Party and the Socialist Left Party had the aim abolishing the cash for care benefit in their party

272 

 Making, Unmaking and Remaking: The Evolution of Nordic Cash for Childcare Schemes

programme. The third coalition partner, the Centre Party – initially a supporter of the reform, valuing “choice” – has been hesitant. But in 2012 a partial displacement of the benefit was effectuated; the benefit was discontinued for two-year-olds.7 Some centreright municipalities have neglected the new rule and continue to provide the benefit for two-year-olds, mostly in the more conservative/religious regions (Ellingsæter, 2012). This reflects the challenge of the “problem of governance” between the central versus the local levels in countries with a high degree of municipal autonomy (see Evers, Lewis & Riedel, 2005). But the minority right-wing government in office after the 2013 election has announced the reintroduction of the benefit for the two-year-olds – that is, a re-layering – supported by an agreement by the Christian Democratic Party and the Liberal Party (Politisk platform, 2013). Moreover, the benefit amount for one-yearolds was substantially increased in 2014.8 The expectation that this rise would lower demand for childcare services failed, however. The Liberal Party has since announced its withdrawal of support. Motivated by the negative effects of the benefit, especially on the use of childcare among minority children families, the party’s aim is to displace the benefit as soon as there is full coverage of childcare services. The right-wing parties together with the Christian Democratic Party have been less concerned by these negative effects; they are holding on to “choice” as their overarching concern. However, the Conservative Youth Party decided in 2014 to aim at the abolishment of the national cash for care benefit and make it a voluntary municipal arrangement. The Centre Party still supports the benefit for one-year-olds, but it should be conditional on one of the parents actually staying at home with the child in the period the benefit is received. The cash for care benefit has had a declining importance to parents, with take up decreasing sharply from 75 per cent in 1999 to 25 per cent in 2011. The decline reflects the strong expansion of places and lower parents’ fees in childcare services. The share of recipients has declined in all groups, but those receiving the benefit have become more homogeneous: Parents with low income and education, immigrants from Asia and Africa and mothers with weak labour market attachment are overrepresented. The impact of the reform on mothers’ employment was negative in the years after the reform (Schøne, 2001; Rønsen, 2009). From the mid-2000s, the employment rate among mothers with children aged one to two started to increase again. The emphasis of criticism among political opponents has shifted somewhat: from a main focus on gender equality to paying more attention to the negative impact on the integration of immigrant mothers and their children (Prop. 1S [2011-2012]). This must be seen in

7 The benefit for the 1-year olds was increased for the 13-18 months olds (NOK, 5000). Either a full or a 50 per cent benefit is paid, the latter for children who attend childcare services less than 20 hours/ week (Prop. 1 S [2011-2012]). 8 NOK 6000/month as of 1 July 2014.



Sweden: Layering, Displacement, Re-Layering 

 273

light of the dramatic reduction in the proportion of benefit recipients among ethnic Norwegian parents, while it remains quite high in certain immigrant groups. The popular support and opposition has been rather stable, and it continues to divide the Norwegian population into two equal parts (Gulbrandsen, 2009).

14.7 Sweden: Layering, Displacement, Re-Layering In Sweden, cash for childcare benefits (vårdnadsbidrag) have also evolved through a long history of political struggle. The question of whether the government should support women’s care work by cash allowances or women’s employment by expanding public childcare services was debated in Sweden after the Second World War (Nyberg, 2010). In the beginning of the 1960s a line of division emerged between those who wanted to strengthen the male breadwinner model and those who wanted a shift towards an individual model with two breadwinners. The policy choice was between a home care allowance and expanding public childcare (Bergqvist et  al., 1999). This division was even present within the governing Social Democratic Party. The Right Wing Party (later the Moderate Party) opposed the idea of a home care allowance; the “natural function of parenthood” should not be supported by the state. However, a decade later views had shifted towards supporting such an allowance, and the Moderates went into an alliance with the Centre party (agrarian base), a strong supporter of the benefit since the 1960s. The principal arguments were freedom of choice and justice in the distribution of state transfers to families. Over the same period the Social Democratic Party developed a firmer negative stance towards it. Hence, since the late 1970s the main priorities among social democrats have been parental leave and childcare services, compared to cash allowances as the main aim of the centre-right (Nyberg, 2010). Throughout the 1980s the centre-right parties intensified their promotion of neoliberal alternatives that stressed choice, decentralization, markets and privatization with regard to childcare (Bergqvist & Nyberg, 2002). After several years of heated debate, a cash-for-care benefit was introduced for the first time in 1994 by a centreright government (Nyberg, 2010). It was part of a political compromise which also included a new gender equality measure, an earmarked “daddy month” in the parental leave arrangement, promoted by the Liberal Party. The government also expanded the social democratic/liberal line of childcare services (Bergqvist & Nyberg, 2002). This was also a process of policy layering with the cash for care benefit added on top of existing childcare services. However, a prompt process of displacement took place later that same year: The benefit was abolished when the social democrats returned to power. Particularly the small Christian Democratic Party has continued the pressure to re-introduce a cash for care benefit; this is one of their top priorities. The current scheme (vårdnadsbidrag) was introduced in 2008 by the majority centre-right coalition that

274 

 Making, Unmaking and Remaking: The Evolution of Nordic Cash for Childcare Schemes

came into power in 2006 (re-elected in 2010). However, this new scheme is voluntary for municipals. In 2011, 37 per cent of all municipalities had introduced a cash for care benefit, mostly in centre-right local governments (Nyberg, 2010; Statistiska Centralbyrån, 2012). The reform’s main rationale was to “increase families’ freedom of choice […] by reducing the national political steering” (Prop. 2007/08:91, p. 17). The benefit is supposed to enable parents to spend more time with their children and to support flexibility in the combination of employment and care. The reform was the outcome of negotiations among the coalition parties, and views differed (Nyberg, 2010): The main protagonist was the small Christian Democratic Party, which wanted the scheme to be obligatory for all municipalities. The Centre Party and the Moderate Party have been in favour of the benefit, while the Liberal Party reluctantly accepted it. The reform this time was also accompanied by a new gender equality measure: a gender equality bonus (tax break) in the parental leave system. The Social Democratic Party, the Left Party and the Green Party opposed the reform because women are considered to be the big losers: It is old-fashioned, it does not support equal parenthood and it risks worsening women’s situation in the labour market (Nyberg, 2010). The Christian Democrats want to double up the benefit, but this is not supported by the coalition partners because it contradicts the “work line” and counteracts the integration of immigrant women in the labour market.9 The Moderate Party has not signalled any discontinuation of the reform. But a recent report from a party working group acknowledges that although the cash for care benefit has entailed more flexibility in the transition from parental leave to childcare services, a negative consequence is that the benefit delays labour market entrance for mothers and keeps children away from important learning in childcare services (Nya Moderaterna, 2014). Still, the cash for care benefit has had a marginal influence on parents’ care arrangements; few parents take the benefit. In the municipalities that have introduced the scheme, the proportion of parents that received it sometime during 2011 was 4.7 per cent, representing 2.5 per cent of all Swedish one to two-year-olds (Statistiska Centralbyrån, 2012). Low educated and immigrants are overrepresented (Segendorf & Teljosuo, 2011). Public childcare services were very well developed when the cash for care benefit was introduced, and the benefit has had little impact on childcare usage.

14.8 Nordic Cash for Childcare Schemes: Stability and Instability The current Nordic cash for childcare schemes have evolved over a considerable period of time. The idea of such schemes dates back several decades. The initial reform thrust of the 1960s was founded on the male breadwinner/housewife family ideal of the

9 ‘KD vill dubbla vårdnadsbidrag’, Dagens Nyheter 1.2.2012.



Nordic Cash for Childcare Schemes: Stability and Instability  

 275

industrial age, and proponents to some extent crossed party lines. However, during the 1970s a diverging line emerged between proponents of the traditional family model supporting home based care and a new model based on collective childcare, mothers’ employment and gender equality. Since the 1970s cash for childcare has been promoted by the political centre-right and opposed by the political left. Thus protagonists and antagonists of this particular type of family policy are associated with party cleavages following the ideological commitments of established “party families” to a considerable degree. Still, over time the original justification has adapted to the social reality and political rhetoric of post-industrial society. Universal childcare services are acknowledged as a key welfare state responsibility by all political parties, and the claim for a wage for housewives or mothers is replaced by parental “choice”. The evolution of the schemes shows important differences between the three countries, however. Reform initiatives have been part of the political struggles, resulting in institutional change through layering, (partial) displacement and re-layering processes. The patterns identified demonstrate the usefulness of an analytical framework highlighting the link between party competition across nations and time and different modes of incremental institutional change. The main patterns are summarized in Table 2. Table 14.2: Evolution of cash for childcare schemes and modes of institutional change

Timing legislation Coalition Main cognitive framework Layering Displacement Re-layering

Finland

Norway

Sweden

1985-1990 Left-centre-right Choice + -

1998 Centre-right Choice + Partial (2012) Partial (announced)

1994, 2008 Centre-right Choice + + Full (1994) + Modified (2008)

When cash for care schemes have been instituted, it has been through policy layering – adding on top of existing childcare services. Childcare services represent a vital interest for the large majority of working parents’ organization of everyday life (Ellingsæter, 2014). The political costs for any party proposing cuts in services would be very high. Left-right competition is likely to continue mainly in the struggle over privatization of service delivery (Mahon et al., 2012). Some have labelled the introduction of cash for care schemes in Nordic childcare policies as “model hybridization” (Knijn & Smit, 2009). But a crucial point is that that the resulting hybrid models are products of political struggle and of conflict and/or compromise. This also means that they are vulnerable to shifts and thus instability (Mahoney & Thelen, 2010). Political conflicts surrounding the schemes contain a continuous pressure for change. The three current national schemes rest on different political coalitions.

276 

 Making, Unmaking and Remaking: The Evolution of Nordic Cash for Childcare Schemes

Change in political government coalitions is the main driver of change and source of instability. In Norway and Sweden the cash for care benefit has been subject to partial or full displacement and (announced) re-layering. The political controversies and compromises underscore the continued significance of the left-right cleavage in family policies, but small parties at the political centre may play a central role in Nordic family policies. The influence of confessional parties in negotiations over government coalitions/platforms has been substantial in both Norway and Sweden. Support for cash for care schemes has been at the centre of negotiations. Agrarianbased centre parties have also been scheme proponents. These are rather small parties in Norway and Sweden but have a considerable power base in Finland. Homebased care resonates with the traditional family values and rural interests thought to be crucial to core voters of these parties. Governments may contain both protagonist and opponent parties, and sometimes cash for care reforms have been balanced with new gender equality measures. The contours of three different national trajectories have emerged: one where the cash for childcare has been relatively durable and two more unstable paths. In Finland, a relatively durable scheme has developed. More than a decade ago, Finland’s cash for care benefit reform was seen as moving the country towards a “neofamilist” model (Mahon, 2002). This projected shift has proven quite resilient. The broad political coalition backing the policy has curbed initiatives for change, though a recent government proposal signals that the policy is not unchangeable. Broad popular support and relatively high benefit usage underpin stability, however, making change costly for political actors. Over time a powerful discourse on parental “choice” has been established. By contrast, the Norwegian cash for care scheme is unstable, and its impact in terms of recipients is shrinking due to improved access to childcare services. A centre-left government coalition resulted in only partial displacement of the benefit because full displacement was not supported by all government coalition partners. The present minority right-wing government has announced annulment of the partial abolishment. Parents’ declining take-up rate has not changed popular attitudes, however. A shift of focus, from gender equality to immigrant integration issues, is evident in the political opposition against the cash for care benefit. The Swedish scheme is unstable and marginal with regard to both policy durability and parents’ usage of the benefit. Sweden’s present cash for care scheme is of limited scope and has resulted in very low take-up rates. The minority red-green government in office since 2014 wants to abolish the scheme. Clearly, the magnitude of impact in terms of eligible parents’ take-up rates varies extensively, and the development over time has been very different (Duvander & Ellingsæter, forthcoming). This has to do not only with the different content of the schemes but also with the different timing of reform with regard to mothers’ labour market situation and access to childcare services for the youngest children. Still, the socio-economic profile of those taking the benefit is similar across the three countries: mothers with low incomes, low educational levels and immigrant backgrounds are



Nordic Cash for Childcare Schemes: Stability and Instability  

 277

overrepresented (ibid.). This is also the reason why several national and international expert commissions have proposed that the cash for childcare schemes should be abolished or significantly reformed (NOU, 2009, 2011; OECD, 2005, 2009). A particular concern is that such schemes keep immigrant mothers out of employment and their children out of daycare. Even so, most protagonist parties hold on to cash for care benefits, promoted as parental “choice” – a central family policy value. The centreright seems to have won issue ownership of “choice”, which also is important for sizable groups of voters (Ellingsæter, 2014). This is probably a major reason for why the cash for care benefit is unlikely to disappear from the centre-right policy agenda in the near future.

278 

 References

References Bakken, F., & Myklebø, S. (2010). Kontantstøttens utbredelse og foreldres preferanser for barnetilsyn – en studie av årskullene 1998-2008 og deres foreldre. Oslo: Arbeids- og velferdsdirektoratet. Bergqvist, C., Kuusipalo, J., & Styrkarsdóttir, A. (1999). The debate on childcare policies. C. Bergqvist, A. Borchorst, A.-D. Christensen, V. Ramstedt-Silén, N. C. Raaum & A. Styrkársdóttir (Eds.), Equal democracies? Gender and politics in the Nordic countries. Oslo: Scandinavian University Press. Bergqvist, C., & Nyberg, A. (2002). Welfare state restructuring and child care in Sweden. In S. Michel & R. Mahon (Eds.), Child care policy at the crossroads. New York: Routledge. Duvander, A.-Z., & Cedstrand, S. (2012). Cash for childcare. En översikt av systemen i de nordiske länderna. Stockholm: Nordens Välferdscenter. Duvander, A.-Z. & Ellingsæter, A.L. Cash for childcare schemes in the Nordic welfare states: diverse paths, diverse outcomes. Forthcoming. Ellingsæter, A. L. (2003). The complexity of family policy reform. The case of Norway. European Societies, 4, 419-443. Ellingsæter, A. L. (2007). ‘Old’ and ‘new’ politics of time to care: Three Norwegian reforms. Journal of European Social Policy, 17(1), 49-60. Ellingsæter, A. L. (2012). Cash for childcare: Experiences from Finland, Norway and Sweden. Berlin: Friedrich Ebert Stiftung. Ellingsæter, A. L. (2014). Nordic earner- carer models – Why stability and instability? Journal of Social Policy, 43(3), 555-574. Evers, A., Lewis, J., & Riedel, B. (2005). Developing child-care provision in England and Germany: Problems of governance. Journal of European Social Policy, 15(3), 195-209. Eydal, G. B., & Rostgaard, T. (2011a). Gender equality revisited – Changes in Nordic childcare policies in the 2000s. Social Policy and Administration, 45(2), 161-179. Eydal, G. B., & Rostgaard, T. (2011b). Day-care schemes and cash-for-care at home. In I. V. Gìslason & G.B. Eydal (Eds.), Parental leave, childcare and gender equality in the Nordic countries. Copenhagen: Nordic Council of Ministers. Fleckenstein, T. (2011). The politics of ideas in welfare state transformation: Christian democracy and the reform of family policy in Germany. Social Politics, 18, 543-571. Fleckenstein, T., & Lee, S. C. (2014). The politics of postindustrial social policy: Family policy reforms in Britain, Germany, South Korea and Sweden. Comparative Political Studies, 47(4), 601-630. Giddens, A. (1991). Modernity and self-identity. Cambridge: Polity Press. Greve, B. (2011). Editorial introduction: The Nordic welfare states – revisited. Social Policy & Administration, 45(2), 111-113. Gulbrandsen, L. (2009). The Norwegian Cash-for-Care reform. Changing behaviour and stable attitudes. Nordic Early Childhood Education Research, 2, 17-25. Haataja, A., & Nyberg, A. (2006). Diverging paths? The dual-earner/dual-carer model in Finland and Sweden. In A. L. Ellingsæter & A. Leira (Eds.), Politicising parenthood in Scandinavia. Bristol: Policy Press. Haataja, A., & Valaste, M. (2012). Cutting child home care allowance in Finland? ESPAnet conference, University of Edinburgh, September 6-8 2013. Hall, P. A. (2010). Politics as a process structured in space and time. Annual meeting of the American Political Science Association, Washington DC, 4 September 2010. Häusermann, S., Picot, G., & Geering, D. (2013). Review article: Rethinking party politics and the welfare state – Recent advances in the literature. British Journal of Political Science, 43, 221-240.

References 

 279

Hiilamo, H., & Kangas, O. (2009). Trap for women or freedom to choose? The struggle over cash for child care schemes in Finland and Sweden. Journal of Social Policy, 38, 457-475. Håland, K. N. (2001). Kontantstøtten – et veiskille i norsk familiepolitikk? Trondheim: IFIM. Knijn T., & Smit, A. (2009). Investing, facilitating, or individualizing the reconciliation of work and family life: Three paradigms and ambivalent policies. Social Politics, 16, 484-518. Knudsen, K., & Wærness, K. (2001). Kontant evaluering. Tidsskrift for Velferdsforskning, 4, 252-258. Korpi, W. (2000). Faces of inequality: Gender, class and patterns of inequalities in different types of welfare states. Social Politics, 7, 127-191. Lammi-Taskula, J., & Takala, P. (2009). Finland: Negotiating tripartite compromises. In S. B. Kamerman & P. Moss (Eds.), The politics of parental leave policies. Children, parenting, gender and the labour market (pp. 87-102). Bristol: Policy Press. Leira, A. (1992). Welfare states and working mothers. Cambridge: Cambridge University Press. Lewis, J. (2008). Childcare policies and the politics of choice. The Political Quarterly, 79, 499-507. Mahon, R. (2002). Child care: Toward what kind of “social Europe”? Social Politics, 9, 343-379. Mahon, R., Anttonen, A., Bergqvist, C., Brennan, D., & Hobson, B. (2012). Convergent care regimes? Childcare arrangements in Australia, Canada, Finland and Sweden. Journal of European Social Policy, 22, 419-431. Mahoney, J., & Thelen, K. (2010). A theory of gradual institutional change. In J. Mahoney & K. Thelen (Eds.), Explaining institutional change. Ambiguity, agency, and power (pp. 1-37). Cambridge: Cambridge University Press. Morgan, K. J. (2013). Path shifting of the welfare state. Electoral competition and the expansion of work-family policies in Western Europe. World Politics, 65, 73-115. Morgan, K. J., & Zippel, K. (2003). Paid to care: The origins and effects of care leave policies in Western Europe. Social Politics, 10, 49-85. North, D. (1999). In anticipation of the marriage of political and economic theory. In J. E. Alt, M. Levi & E. Ostrom (Eds.), Competition and cooperation: Conversations with nobelists about economic and political science (pp. 314-317). New York: Russell Sage Foundation. NOU (2009). Fordelingsutvalget, NOU 2009: 10. Oslo. NOU (2011). Velferd og migrasjon. Den norske modellens framtid, NOU 2011:7. Oslo. Nya Moderaterna (2014). Välfärdspolitik som förbättrar barns uppväxtvillkor. Slutrapport från arbetsgruppen «Lika möjligheter för alla barn. Stockholm. Nyberg, A. (2010). Cash for childcare schemes in Sweden: History, political contradictions and recent developments. In J. Sipilä, K. Repo & T. Rissanen (Eds.), Cash-for-childcare. The consequences for caring mothers. Cheltenham: Edward Elgar. OECD (2005). Babies and bosses – Reconciling work and family life. Canada, Finland, Sweden and the United Kingdom. Paris. OECD (2009). Jobs for immigrants: Labour market integration in Norway. Paris: OECD. OECD (2010). Gender Brief. Paris. Retrieved from www.oecd.org/els/social Pierson, P. (2004). Politics in time. History, institutions and social analysis. Princeton: Princeton University Press. Politisk plattform for regjering utgått av Høyre og Fremskrittspartiet, Sundvollen 7.10.2013. Prop. 2007/8:91. Vårdnadsbidrag – familjpolitisk reform. Stockholm: Socialdepartementet. Prop. 1 S (2011-2012). Oslo: Barne- likestillings- og inkluderingsdepartementet. Rantalaiho, M. (2009). Kvoter, valgfrihet, fleksibilitet. Oslo: NIKK. Rantalaiho, M. (2010). Rationalities of cash-for-childcare: The Nordic case. In J. Sipilä, K. Repo & T. Rissanen (Eds.), Cash-for-childcare. The consequences for caring mothers. Cheltenham: Edward Elgar. Repo, K. (2010). Finnish child home care allowance – Users’ perspectives and perceptions. In J. Sipilä, K. Repo & T. Rissanen (Eds.), Cash-for-childcare. The consequences for caring mothers. Cheltenham: Edward Elgar.

280 

 References

Rønsen, M. (2009). Long-term effects of cash for childcare on mothers’ labour supply. Labour, 23, 507-533. Salmi, M. (2006). Parental choice and the passion for equality in Finland. In A. L Ellingsæter & A. Leira (Eds.), Politicising parenthood in Scandinavia. Bristol: Policy Press. Salmi, M., & Lammi-Taskula, J. (2013). Finland country note. In P. Moss (Ed.), International review of leave policies and research 2013. Retrieved from http://leavenetwork.org/lp_and_reports/ Schøne, P. (2004). Labour supply effects of a cash-for-care subsidy. Journal of Population Economics, 17, 703-727. Seeleib-Kaiser, M. (2010). Socio-economic change, party competition and intra-party conflict: The family policy of the grand coalition. German Politics, 19, 416-428. Segendorf, Å.O., & Teljosuo, T. (2011). Sysselsätting för invandrare – en ESO-rapport om arbetsmarknadsintegration. Stockholm: Finansdepartementet. Sipilä, J., Repo, K., & Rissanen, T. (Eds.) (2010). Cash-for-childcare. The consequences for caring mothers. Cheltenham: Edward Elgar. Statistiska Centralbyrån (2012). Nyttjande av kommunalt vårdnadsbidrag. Stockholm. Vollset, G. (2011). Familiepolitikkens historie – 1970 til 2000. Oslo: Nova. Weyland, K. (2008). Toward a new theory of institutional change. World Politics, 60, 281-314.

PART IV: Afterword

Fredrik Engelstad

15 Conflict, Compromise, Cooperation Concluding Reflections In her Varieties of Liberalization and the New Politics of Social Solidarity (2014) Kathleen Thelen compares the tracks of change in the Continental and the Scandinavian varieties of Coordinated Market Economies. In welfare policy studies the systematic difference between the two parts of Europe was first drawn up by Gøsta Esping-Andersen (1990), and it has intuitively been applied to draw a similar demarcation line in political economy. What Thelen demonstrates empirically – far beyond mere assumption – is not only that these are two different areas of political economy, but that they develop in different directions; the Continental countries are moving towards increasing dualization of the labour market, whereas Scandinavian economies have been able to retain their solidaristic character. This is explained by two salient features. First, flexibility. By not clinging to given institutional structures, but modifying core institutions such as labour market policies, bargaining systems and vocational training, Scandinavian countries have been able to preserve a high degree of equality. Second, the tripartite power constellation of strong employer associations, strong trade union federations, and a strong and liberal state. This institutionalized pattern of conflict and cooperation is probably the most important precondition for a form of flexibility that does not make the working classes the main losers in economic changes. Thelen’s analyses invite further reflections, and the preceding case studies in this book may concretize, complement, expand, and to some extent modify her conclusions. Covering primarily one country – Norway as the most extreme example of neo-corporatism – they have the advantage of concentration and depth; all of the cases cover Norway and about half of them keep a comparative link to the rest of Scandinavia. The cases in no way give a complete coverage of Norwegian society; but taken as a whole they add broad perspectives and salient details to the general picture, which is necessarily painted with broad strokes. The present studies indeed confirm the interrelation of solidarism with flexibility and tripartite bargaining structures. But to be understood, the process needs further empirical specification.

15.1 Flexibility – Balancing Conflict and Contract Historically, the critical event was the Basic Agreement of 1935. But how could it happen at all that two parties in deep conflict enter into a contract that constitutes an agreement to disagree? A standard point of departure would take the form of a Prisoner’s Dilemma problem; reaching an agreement is optimal for both parties, but how to get it established if neither party has a short term interest in yielding to the © 2015 Fredrik Engelstad This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivs 3.0 License.



Emergence and Expansion of the Tripartism 

 283

other? As in many other cases, the problem was solved by state regulation; not in the conventional form of a legislative dictate, but by threats of state intervention, something that both parties feared more than establishing in common a modus operandi. This solution, in turn, rests on three preconditions. The parties must acknowledge that they have a common interest. The threat from the state must be credible. And a possible solution must be sketched which presents itself as neutral to the short term interests of the parties. All three conditions were fulfilled in the Norwegian case. After the dramatic class conflicts of the 1920s followed by the economic collapse in 1929, both labour market parties realized they could not be the winner of a total conflict. The state was credible due to its strong tradition of political intervention into the economy already from the mid-19th century, the so called Norwegian System (Seip 1968). And a solution was already preformed in the 1899 agreement in Denmark, albeit a weaker agreement than what was to become the Norwegian one (Due & Madsen, 2010, Martin & Swank, 2012). These background factors channelled the Prisoners’ Dilemma situation into that of an Assurance Game, and initiated a long period of relatively stable cooperation between the parties. This process emphasizes ambiguity as an essential trait of a large subtype of institutions. By institutionalizing conflict they do not regulate behaviour in a direct way, but establish arenas for conflict, not least class conflict (Dahrendorf 1959). Even if willingness to accept common rules and settle compromises develops a viable relationship between opposing parties, it does not neutralize conflict, only gives it a different form. In institutionalized power games, conflicts are located at three different levels. Primarily they are played out directly between opposing parties in regular bargaining. Simultaneously, conflict is fought out more indirectly over the regulative rules of the game, i.e. the regular mode of functioning of a given institution. Moreover, over long time stretches, conflict is also played out over constitutive rules, i.e. the basic form and justification of the institution. To the degree that flexibility is the hallmark of institutions, it has as an implication that conflicts are fought out and changes are taking place on all three levels. It is a commonplace that action entails unintended consequences. The complexity of conflict patterns makes outcomes even less predictable. Due to the short term character of bargaining processes, in a broader perspective they take on the form of trial-and-error, leading up to compromises and arrangements imagined by none of the actors in advance, as illustrated by several contributions here.

15.2 Emergence and Expansion of the Tripartism Even though the original Basic Agreement only covered a limited sector – organized companies and employees in the private sector – it was to have very large consequences. First of all, the relationship between the labour market parties was strengthened

284 

 Conflict, Compromise, Cooperation Concluding Reflections

and reinforced. After three decades of renegotiation, further specification and adjustments, a significant enlargement of the agreement was introduced in 1966. New elements were added, establishing modes of codetermination for employees, linked to rights to information and participation in organizational development. This reflects growing trust in the employer-employee relationship, but it also mirrors the general observation that powerful bargaining parties see it in their interest to have strong adversaries, because it guarantees stable and predictable relations. This is illustrated by the diffusion of basic agreements to other sectors of the economy, such as the service sector, the financial sector and the public sector. Thus the Basic Agreement was virtually generalized to the Norwegian economy as a whole. Albeit taking an active role in pressuring for an agreement between the labour market parties, the state remained basically outside the bargaining process. But the reinforcement of the Basic Agreement, along with scientific development of economics, gradually led to a change in the role of the state. Most important was the emergence of tripartite wage bargaining on the national level from the mid-1960. Actually, the group most strongly pressing for this reform was the employers (Petersen, 1975, after Reiersen, 2012). In the next round, the state was more deeply involved, gradually taking on a double role. It greatly expanded its apparatus for facilitating the bargaining process, creating meeting places for discussion and setting up neutral, advisory committees of economic experts. From time to time the state would also take a more directly active role in negotiations, by introducing political reforms to induce the parties to reach an agreement. Moreover, without the reinforcement and expansion of the Basic Agreement, the ensuing legislation on industrial democracy, both at the job level and the enterprise level, would hardly have been possible. The effects of the Basic Agreement may also be measured by its traces in the social fabric of the welfare state as a whole. The class compromise and the ensuing culture of bargaining meant a strengthening of the political positions of the labour market federations, making them the most powerful organizations outside the purely political system, with open channels into both the state bureaucracy and national politics. Here, two parallel strands become relevant. The welfare state is not only a set of benefits and provisions to citizens, but just as much an employment system. Thus, the labour market parties have strong interests and are directly involved in the organizational design and functioning of the welfare state apparatus. In addition, the welfare state has numerous direct and indirect effects on the functioning of working life, whether it be educational systems, organizing of sick pay insurance or pension plans, just to name a few. It follows that the tripartite constellation of trade unions, employers and the state, which has its origins in the labour market, is generalized to core parts of the welfare state, granting them significant bargaining power in the political realm as well.



Reform as Routine  

 285

15.3 Reform as Routine The catchphrase Reform as routine (Brunsson, 2009), underlines that organizational flexibility is a necessity elicited by constant processes of change. This formulation also aptly catches neo-corporatist society as a whole. The close interplay of politics and broad interest organizations becomes a driver in large reform issues such as general pension plans, social security schemes, health care, or vocational training, but also in numerous minor adjustments of the welfare system. In addition to formalized reforms it also opens for the occurrence of compromises such as the institution of temporary layoffs, which emerged by informal cooperation between employee and employer associations, while at the same time being upheld by adjustment of rules allowing state funding through allocation of unemployment benefits. It is a common assumption in the literature on welfare state reform that left wing parties in government constitute a crucial factor to bringing about such reforms. This, however, is only half of the story. In Norway the close cooperation of the Labour Party and the major trade union federation LO undoubtedly was beneficial to the reinforcement of the Basic Agreement and the development of trust between the labour market parties. But the decisive impulse from the state to the establishment of the Basic Agreement came when the liberal party Venstre was in government. In the more than four decades since 1970, minority governments have been the main rule in Norway; which means that welfare state reforms were dependent on support from both the left and the right. The motor of continuous reform is composed of several elements. The strength of organized labour is crucial, but so is a broad variety of social movements able to engage with the state in order to achieve political reforms. Another is the willingness to achieve compromise, manifested in a tradition of what has been termed “secondary elite compromises” (Gulbrandsen et al, 2002, p. 37-42), concerning issues such as language politics, religious pluralism, equality of gender, in addition to labour market compromises resting on liberty of organizing and rights to private property. A third element is a strong, but nevertheless liberal state, with a fairly efficient and incorrupt bureaucracy. And finally, a firm cultural foundation for the universalization of norms and values of citizenship, operationalized in concepts of equality, equity, justice, and fairness. Such expansive reform developments are not well explained by the leading theories of institutional change. The theory of path dependency is basically focused on equilibria, interrupted by external shocks. It describes how institutional traits are preserved and reinforced over time by alternative options becoming increasingly costly. This would explain why the Basic Agreement was established as a response to economic crisis, kept intact and to some extent revised in the following years. What remain unexplained are the subsequent establishment of the bargaining system and the expansion of the welfare state.

286 

 Conflict, Compromise, Cooperation Concluding Reflections

The theory of aggregate change, on the other hand, has mostly concentrated on the redefinition of institutional arrangements by external factors, such as cultural reinterpretation (conversion), and changes in the relevance of institutions by contextual effects (drift). Equally important, at least, but less explored, is politically induced change bringing forth various forms of layering. A point demonstrated in this volume is that layering not only appears as the result of given reforms, but becomes a source of complex dynamic processes after the reform is implemented. Layering may resolve problems, but equally well create new dynamics; new elements may drive the old ones out, or on the contrary be rejected, or their combination may create insuperable tensions over a long time. If reforms become routine, the occurrence of layering increases; thus the theme of institutional complementarities, a prominent issue in the Varieties of Capitalism literature (Hall & Soskice, 2001, p. 17-24), is brought up. Securing compatibility between institutional traits becomes crucial in continuous reform processes, not only within a given institution, but between institutions as well. There is no doubt that this contributes to the stability of the overall institutional structure. At the same time, Thelen (2012) rightly points out that the effects on stability are easily exaggerated. This point may be taken a step further. Increases of reform entail increases in complexity. The outcome is not only increased stability, but also increases in tensions and unpredictability. One source of unpredictability is the problem of handling complex sets of rules in a consistent way. Others are linked to unintended and unpredicted tensions between institutions, as e.g. in the effects of child care allowances on motivating immigrant mothers to stay out of paid work, and thereby reinforcing their social isolation.

15.4 Norms in Institutional Change In a now classical definition of institutions, Richard Scott juxtaposes what he characterizes as three elements: “cultural-cognitive, normative and regulative elements” ([1995] 2008, p.48). But even though these aspects necessarily are amalgamated, in established theory the normative aspect of institutional change seems to go unnoticed more often than not. In the path dependence approach norms are accorded a minor role. In the conception of aggregate effects the impact of norms is neutralized by references to power resource theory, where normative issues are explicitly played down (Korpi 1985). One reason may be found in an ambiguity in the concept of institution, as coined by Scott. In a general sense institutions are normative already by the fact that they are regulative rules, made up by prescriptions and/or proscriptions. These rules are resting on common normative grounds. This is also how they are envisaged by Scott; in his view norms are of a unified, and thus so to say conservative, nature. However, in many cases institutions are normative also in a different sense, because they function



Norms in Institutional Change 

 287

as frameworks for struggles over normative issues, over which norms are relevant and how they are to be interpreted. In this sense, norms form a nested structure, where regulative rules are shaping struggles over interpretation and application of norms. In bargaining over norms, they do not take the character of a unequivocal rule, but as a continuum of specific interpretations that the parties struggle over, but nevertheless recognize as legitimate. In turn, reinterpretation of norms entails institutional change. A telling counter-example, of deinstitutionalization due to transgression of legitimate norm conflict, stems from Sweden in the early 1980s, when the social democratic government introduced the Wage Earner Funds. They were established by channelling significant parts of profits, and thereby of investment decisions, from enterprises into “socialized” investment funds run by political appointees. The funds constituted a breach with the established modus vivendi and created a deep split between the employers on the one side and the social democrats and the trade union movement on the other. As a long term effect a significant part of the tripartite structure in Sweden was demolished. When a centre-right government came to power in 1991, the funds were dismantled, but trust and cooperation between the Swedish labour market parties never again reached the earlier level. Rather than assuming a contradiction between power and norms, it is important to acknowledge how norms are intertwined with the exertion of power. Elster (1989) has already pointed out that in wage bargaining, strategic and normative aspects exist side by side and both are activated in the process. The two aspects are difficult to disentangle; norms are often invoked strategically, but that does not imply they are ephemeral. On a different level, the same is true for negotiations over welfare state arrangements. The preceding chapters amply demonstrate how norms, norm conflicts and changes in the interpretation of norms, contribute to shaping institutions. Notions of equality, equity, justice and fairness are quickly brought up in politically induced reforms. These concepts have a strong emotional appeal, while at the same time they are ineradicably ambiguous. They continuously engender deliberation and struggles over meaning, interpretation and reinterpretation. In addition to norms introduced directly into bargaining processes, another set of norms has emerged on the structure of wage formation. The tripartite wage bargaining system has inspired a set of normative claims and counter claims which is specific to neo-corporatist society; concerning not only the justifiability of outcomes, but of bargaining process as well. What kind of bargaining system is more just, emerges as a separate field of conflict: a pure free market model, a state governed redistributive system, or a large scale tripartite bargaining process where norms are represented by the social partners, rather than the state? Changes in the extension and interpretation of property rights have developed by trial-and-error over several decades. Trade unions, employers, and the state alike have changed their position in several rounds, and reforms have been modified after their adoption. Rules for the composition of company boards were set by legislation, with the aim of extending democratic rights to the economic sphere. But power also

288 

 Conflict, Compromise, Cooperation Concluding Reflections

rests in the invocation of norms. Employers could refer to the norms of property rights in order to keep the changes within certain limits, preventing the extension of employee representation on boards to 50 per cent. The more recent reform of gender quotas to boards was also assumed by employers to be an attack on property rights. But here the normative conflict was mitigated by other sets of shared norms. In this case employer representatives were bound by their adherence in public to lifting gender discrimination; they disagreed on the means, not on the ends. Sickness pay is another field where norms turn out to have a pivotal role. The blockage of changes to the present system of full wage compensation for sickness absence is due to the power of the coalition of trade unions and employer associations. But their position would hardly enjoy the broad support it does, without the widespread normative concern in the public for citizens who have had the misfortune to fall ill. A related case is that of innovative changes in social services, enacted by employees, but with main reference to the quality of the services for their users. That professions refer to the welfare of their clients to secure advantages for themselves is an old story, but from time to time it happens to be true. If not, the argument would not have any impact. Finally, the cash for childcare programs also reflect normative ambiguities. One norm of equality amounts to equality between users and non-users of heavily subsidised kindergartens, and hence is an argument for monetary compensation to non-users. A counter-norm would relate to citizenship, viewing preschool socialization on a par with regular schools. If so, it does not make sense to demand compensation for staying away from kindergartens, in the same way as it would be strange to expect monetary compensation for keeping children out of school.

15.5 Sustainability of Egalitarian Capitalism? Few observers, even among sceptics, would deny the success of the Nordic model up to now. The big question concerns its long term sustainability; there is no lack of worries, even among adherents to the model, whether they are on the political left or right. Two studies discussing the future of the Nordic model were recently published in Norway, by the liberal think tank Civita (Fasting, 2014) and by Fafo Institute for Labour and Social Research (Dølvik et al, 2014). Both contributions are timely and relevant; by reflecting different theoretical emphasis (“Rational Choice” vs. “Industrial Relations”) and political orientations. Very roughly: The Civita diagnosis has as its main focuses the challenges to the welfare state due to international economic trends and to the ageing of the population. It calls for increased flexibility in the public sector by increased privatization and in the private sector by more flexible working conditions. The Fafo analysis does not deny the need for flexibility, but is more worried about attacks on the regularities of working life, along with growing inequalities of income and welfare, as well as decline in unionization rates. The Civita



Sustainability of Egalitarian Capitalism? 

 289

report advocates tax cuts and expanding the scope of action of employers; the Fafo report sets its hopes on continued public organization of welfare state services and on a strategic role for trade unions. Both views will certainly have a prominent position in negotiations over future reforms, even if coming compromises takes other forms. None of the two studies, however, touch more than in the passing on institutional structure; i.e. on the preconditions for these bargaining processes. Both seem to assume that their proposals for reforms can be realized without the basic institutional structure coming under pressure. There may be good reasons for the assumption of institutional stability; the degree of complexity and institutional complementarity gives neo-corporatist society a high degree of resilience (Pierson, 2001). In Norway the legitimacy of the welfare state remains unchallenged both on the left and the right, not least due to is functional success over several decades. Welfare state issues have for three decades held the highest priority among the voters (Karlsen, 2009). Numerous group interests are backed by strong norms of equity in distribution and public responsibility for the welfare of all. This means that any reform of importance requires long term deliberation, as exemplified by the ten years of bargaining process over the Norwegian pension reform. On the other hand, Thelen (2012) points out that even if reforms are put in place, seemingly stable institutions may be altered through processes of drift. In the long run the existing structure may change character; if a tipping point is reached the framework for negotiations may be undermined. Maintaining structures based on norms of equality and solidarity requires flexible readjustment; acknowledging the new constellation and being able to act upon it. Some of the challenges envisaged by the two studies may be discussed in this light; to which degree institutions are sustained by path dependency or threatened by drift. At any rate, major challenges will generate strong conflicts. The main question, however, is to which degree basic institutional structures will be affected. One obvious challenge to the Nordic model concerns possible threats from international trade cycles. This is particularly sensitive to Norway, where the economy is closely linked to the oil price on the world market. Long term declines in oil prices will put pressures both on public budgets and on the private sector economy. But oil is not all there is to Norwegian economic achievements. Like Norway, all the Nordic countries have very open economies, and are accordingly affected by serious oscillations in the world markets. This might be a source of drift. However, the need to meet these challenges was a major impetus to the establishment of tripartism as a part of robust economic policies. Looking back on the effects of the financial crisis in 2008 reveals that the Scandinavian countries were the most able to cope with the crisis; the institutional structure basically remained unaffected. Relatively, Norway will be hit harder than the rest of the Nordic countries by contraction in the oil industry, but this may be balanced to a great extent by large economic reserves in combination with a high degree of flexibility in the economy. In setbacks of a similar severity as the

290 

 Conflict, Compromise, Cooperation Concluding Reflections

finance crisis, it seems more probable that the legitimacy of the Nordic model will be reinforced, as happened then, rather than the opposite. The welfare state is not so much threatened by the world economy as by internal forces, Paul Pierson (2001) has pointed out. An ageing population is undoubtedly a serious internal challenge to the welfare state; which invites confrontations between interest groups to a higher degree than instabilities in the world markets. But even if aging is a form of drift, the long run consequences are well known; a range of political measures are available to cope with it. The example of the successful bargaining over the pension reform is an indication that the institutional framework is robust, and will produce acceptable solutions. At the same time, the Norwegian pension reform is still unfinished, due to lacking integration of public sector pensions into the general system. Negotiations to come may well turn out to have more in common with the stalemate over sickness allowances than with the successful bargaining over pensions in the private sector. Possible changes in the position of trade unions go to the heart of the Nordic model. Without continuous presence of trade unions in policy formation, further adjustments and developments of the welfare state may stagnate. Here too is an instance of drift. In the last decade the unionization rate has shown a significant decrease in Scandinavia as a whole; declining by 10 or more percentage points in Sweden and Denmark over less than two decades, compared to 5 percentage points in Norway in the same period (Dølvik et al, 2014, p. 75). An obvious reason for the decline in unionization is changes in the occupational structure, away from the blue collar jobs that traditionally has constituted the core of the trade union movement. Additionally, in Sweden and Denmark the decline is caused by a partial dismantling of the system of channelling unemployment benefits through the trade unions. Still, around two thirds of the total number of employees in these countries is unionized, in Norway more than a half. According to power resource theory, declining membership means a loss of power. But from an institutionalized perspective the matter may look differently. The major power base of unions rests in their placement in the dense networks around national wage bargaining, and their role as parties to be consulted on a broad set of issues in current political decision-making. The strong position of unions is not guaranteed eternal life, but they enjoy the advantage of path dependence. Marginal decline in membership does not justify changes in the system of consultation and participation of interest groups, which will be formidable. However, in the case of a steady decline in membership, albeit slow, a tipping point will be reached where alternative institutionalization of bargaining will be demanded. Finally, two cases of drift, which are not easily countered by political action or mitigated by political reform, stand out. One is linked to what has been termed the revolt of middle classes (Rothstein, 1998, after Thelen, 1999). Norway has seen a tremendous growth of in average incomes over the last decades. At the same time economic inequalities have been growing, mostly by increased incomes on capital in the very highest income brackets, while the wage structure remains relatively



Sustainability of Egalitarian Capitalism? 

 291

compressed. The upper parts of the middle classes enjoy relatively low wage incomes compared to their European counterparts, but carry a relatively heavy tax burden. Their protests may take two directions: (i) breaking out of the compromises on income formation by pressuring for higher incomes, and (ii) demanding changes in the welfare state provisions, by replacing a system encompassing all with universal minimal standards combined with extra benefits funded by private contributions. The latter case may also be seen as an example of conversion, in that the notion of universalism may be taken to cover both systems, even though the conceptual content differs significantly between them. One of the cases in the book indicates a second challenge, concerning the effects of immigration coupled to problems of control in the labour market. These processes have been effective for at least two decades, but gained additional force by the inclusion of Eastern European countries in the EU in the 2000s. Norwegian trade unions have insisted on equal of wages and working conditions for immigrant workers as for Norwegian citizens, with some success (Hardy, Eldring & Schulten, 2012). Nevertheless, tendencies toward a dualization of the labour market are becoming visible in industries such as cleaning and construction. Significant groups of employers and immigrants may see advantages in labour contracts well below standard level, something that will threaten standard working life regulation. Transaction costs of upholding regulations may be very high, and forms of control quite at odds with well-established forms of labour inspection based on delegation and trust. Initially these problems will be of a local nature, but with diffusion to other branches and industries, dualization may become ingrained and appear as a regular feature of the labour market. Common to these challenges is that they are not resolved by well-established bargaining processes or common types of political reform. How far they will eat into the welfare state fabric is difficult to predict. Coping with them will undoubtedly require flexibility but, even more, considerable amounts of institutional innovation.

292 

 References

References Brunsson, N. (2009). Reform as Routine. Oxford: Oxford University Press. Dahrfendorf, R. (1959). Class and Class Conflict in Industrial Society. Palo Alto: Stanford University Press. Due, J. & Madsen, J.S. (2010). Septermberforliget og udviklingen af den danske aftalemodel. In T. Bergh (Ed.), Avtalt spill. 75 år Hovedavtalen LO-NHO. Oslo: LO og NHO. Dølvik, J. E., Fløtten, T., Hippe, J. M., & Jordfald, B. (2014). Den nordiske modellen mot 2030. Et nytt kapittel? Fafo-rapport 2014:46. Oslo: Fafo. Elster, J. (1989). Cement of Society. Cambridge: Cambridge University Press. Esping-Andersen, G. (1990). Three Worlds of Welfare Capitalism. Princeton: Princeton University Press. Fasting, M. (2014). Har vi råd til fremtiden? Perspektivmeldingens utfordringer. Oslo: Civita. Gulbrandsen, T., Engelstad, F., Klausen, T. B., Skjeie, H., Teigen, M., & Østerud, Ø. (2002). Norske makteliter. Oslo: Gyldendal Akademisk, Hardy, J., Eldring, L. & Schulten, T. (2012). Trade union responses to migrant workers from the ‘new Europe’: A three sector comparison in the UK, Norway and Germany. European Journal of Industrial Relations, 18, 347-363. Karlsen, R. (2009) ‘På dagsorden.’ In A. Bay, A. W. Pedersen & J. Saglie. (Eds.) Når velferd blir politikk. Partier, organisasjoner og opinion. Oslo: Abstrakt. Korpi, W. (1985). Power Resources Approach vs. Action and Conflict: On Causal and Intentional Explanations in the Study of Power. Sociological Theory 3(2), 31-45. Martin, C. J., & Swank, D. (2012). The Political Construction of Business Interests. Cambridge: Cambridge University Press. Petersen, E. (1950). Norsk Arbeidsgiverforening 1900–1950. Oslo: Norsk Arbeidsgiverforening. Reiersen, J. (2012). Arbeidsgiverne og den nordiske modellen. Magma, 4/12: 32-41. Rothstein, B. (1998). Just Institutions Matter. The Moral and Political Logic of the Universal Welfare State. New York: Cambridge University Press. Scott, W. R. (2008). Institutions and Organizations. 3rd edition. Thousand Oaks: Sage. Seip, J. A. (1968). «Det norske system» i den økonomiske liberalismens tid (1850-1870). In J. A. Seip. Tanke og handling i norsk historie. Oslo: Gyldendal. Thelen, K. (1999). Historical Institutionalism in Comparative Politics. Annual Review of Political Science, 2, 369-404 Thelen, K. (2012). Beyond Comparative Statics: Historical Institutional Approaches to Stability and Change in the Political Economy of Labor. In G. Morgan, J. Campbell, C. Crouch & O. K. Pedersen (Eds.), The Oxford Handbook of Comparative Institutional Analysis (pp. 41-62). Oxford: Oxford University Press. Thelen, K. (2014). Varieties of Liberalization and the New Politics of Social Solidarity. Cambridge: Cambridge University Press.

About the Authors Anne Lise Ellingsæter Professor of sociology, University of Oslo Fredrik Engelstad Professor em. of sociology, University of Oslo Eivind Falkum Senior Researcher, Work Research Institute, Oslo Anniken Hagelund Associate Professor in sociology, University of Oslo Inger Marie Hagen Senior Researcher, Fafo Institute for Labour and Social Research, Oslo Cathrine Holst Senior Researcher, Arena Centre for European Studies, University of Oslo Cathie Jo Martin Professor of political science, Boston University Heidi Nicolaisen Senior Researcher, The Norwegian Labour and Welfare Administration Dag Olberg Senior Researcher, Fafo Institute for Labour and Social Research, Oslo Axel West Pedersen Senior Researcher, Institute for Social Research, Oslo Jørgen Svalund Senior Researcher, Fafo Institute for Labour and Social Research, Oslo Mari Teigen Research Director, Institute for Social Research, Oslo Sissel Trygstad Research Director, Fafo Institute for Labour and Social Research, Oslo Mia Vabø Senior Researcher, Norwegian Social Research (NOVA), Oslo

List of Figures Figure 6.1. Proportion of women on the boards of Public Limited Companies, Norway. 2002 – 2013. Source: Statistics Norway. 100 Figure 8.1: The Norwegian temporary layoff institution, distribution of costs. 143 Figure 8.2: Regulatory changes in temporary layoffs. 144 Figure 8.3: Temporary layoffs in the period August 2008 – December 2012 (Source: based on figures from The Norwegian Labour and Welfare Administration NAV). 146 Figure 10.1: Model of the degree of decent working conditions. Source: Trygstad et al., 2011. 187 Figure 10.2: To what extent do you believe that the approval scheme for cleaning companies and the requirement to hold ID cards will improve the conditions in the industry? 193 Figure 10.3: Has extension of the collective agreements wage level to non-unionized workers had an impact on the wage level in your company? 194

List of Tables Table 2.1: Features of acute and incremental change 21 Table 2.2: Distributive bargaining versus inclusive bargaining 24 Table 5.1: Percentage of companies with employee representatives at the board 87 Table 6.1: Types of positive action/quota procedures according to social fields 102 Table 9.1: Types of measure used by company and industry 164 Table 9.2: Types of measure by company and country 170 Table 10.1: Quality of working conditions (WC) and different types of regulation 182 Table 12.1: A summary of the key variables in a multi-perspective analysis of two policy reform cases 237 Table 14.1: Current cash for childcare schemes in Finland, Norway and Sweden 267 Table 14.2: Evolution of cash for childcare schemes and modes of institutional change

275

Index apprenticeship 25, 201-214 asset specificities 64 bargaining 3-5, 7, 10-11, 13-14, 23-24, 28, 31, 33, 38, 40, 44-45, 52-53, 60, 62, 68, 70, 122-123, 125-130, 139, 141, 150, 154-155, 158, 161, 172, 175, 178-182, 185, 191, 196, 200, 214, 227, 230, 238, 243, 282-285, 287, 289-291 Basic Agreement 1, 3, 4, 8, 11, 26, 37, 40-42, 44, 46, 52, 56-57, 61-63, 65-74, 80-81, 83-85, 91-93, 137, 141-142, 144, 148, 151-152, 162, 282-285 Blegen Committee 210 board of directors 6, 11, 37, 40, 45-46, 52, 78, 81, 116 cash for childcare schemes 12, 262-263, 265-267, 274-275, 277-279 childcare services 262-263, 266-276 cleaning 5, 13, 179-183, 185-198, 291 coalitions 1, 6, 20, 22, 29, 36, 60, 136, 138, 149-150, 153-154, 263-264, 275-276 collective bargaining 3, 11, 13, 28, 70, 125-129, 139, 141, 155, 158, 179-182, 196, 200 Companies Act 37, 45-49, 51, 81, 86, 92, 94, 99, 103, 109 Company Assembly 8, 11-12, 47-48, 51, 69, 77-79, 81-95 conversion 9, 20, 28, 37, 45, 48, 57, 65, 79, 196, 222, 236, 244-245, 247, 257, 265, 286, 291 coordinative discourse 12, 234, 237, 238 crisis 3, 7-8, 13-17, 23, 69, 136-140, 143, 146-149, 153, 155-157, 159, 161-163, 165, 168-169, 172-174, 178, 220, 230, 237, 244, 265, 269, 285, 289-290 decentralization 247-248, 250-251, 257, 273 democracy 1, 3-4, 6, 11, 15-18, 22, 29, 32-33, 40, 45-47, 53, 59-60, 64, 66-70, 76-82, 84, 86, 89, 91-93, 95, 103-104, 107-108, 115, 125, 127, 129-130, 178, 243, 247, 258-259, 264-265, 270, 278, 284 Denmark 3-5, 28, 33, 37, 54, 95, 140-142, 153, 158, 161-163, 169-171, 175, 177, 201-202, 262-263, 283, 290 deregulation 99, 104-105, 112, 115, 188, 199 discursive institutionalism 9, 97, 116, 241, 245, 258, 260

dismissal regulation 26, 157-158, 162, 166-167, 168, 173 displacement 2, 9, 37-38, 42, 44, 48, 51, 57-59, 63, 183, 196, 222, 265, 269-270, 272-273, 275-276 drift 9, 12, 20, 29, 37, 57, 93-94, 136-138, 147, 151, 154, 183, 188, 190, 192, 212, 222, 232, 244-245, 247, 254, 257, 265, 286, 289-290 dualization 179-181, 183, 188, 191-192, 195, 197-198, 282, 291 dual system 202 Dynamic Change of Institutions 56 early retirement 163, 166-168, 171, 173, 225, 226, 227, 229, 230, 235, 236 earner-carer model 266, 278 electoral politics 264 employee representation on boards 50, 69, 288 employer associations 39, 59, 179, 181, 282, 285, 288 employment relations 3, 36, 44, 139, 152, 179, 180, 183, 199-200 equal pay 10, 26, 118-131, 133 Finland 12, 21, 36, 60, 96, 140-141, 158, 161, 163, 169, 171, 173-174, 262, 266-268, 270, 276, 278-280 fiscal sustainability 225 gender 1, 8, 10, 17, 30, 37, 48-53, 87, 96-101, 103-119, 122-124, 126-127, 131-133, 262, 266, 268, 271-276, 278-279, 285, 288 gender equality 1, 17, 49, 96-98, 101, 103-107, 109-110, 112-114, 116-117, 123, 126, 133, 262, 266, 268, 271-276, 278 gender quotas 8, 10, 17, 37, 48-49, 51-53, 96-101, 103-116, 288 home care 7, 21, 242, 245, 247-248, 250-252, 254-258, 261, 268-270, 273, 278-279 IA-agreement 224, 228, 232, 234, 236, 238-239 incremental change 1, 6, 20-22, 57-58, 232, 244, 245, 257 industrial democracy 40, 45-47, 53, 67, 77, 80-82, 86, 92, 95, 103-104, 108, 178, 284 industrial relations 17-18, 27-30, 32, 36, 54, 56, 64, 75, 95, 113, 133, 139, 141-142, 156-157, 177-178, 181, 184, 188, 190, 198-200, 213, 240, 288, 292 institutional change 1, 2, 3, 6-12, 15-23, 28-31, 33-34, 37-38, 43-44, 46, 48, 51-53, 55-58,

Index  66, 68-70, 73, 75, 77, 92-93, 95, 97, 116-117, 132-133, 135, 137-138, 147, 155, 157, 160, 176-178, 181-184, 198-199, 202, 212, 214, 216, 220, 222, 224, 236, 239-241, 244, 260, 262-265, 275, 279-280, 285-287 institutional design 56-57, 59, 64, 66, 71, 73 institutional matrix 58, 73-74 interest construction 138, 150 justice 11, 21, 25, 39, 103, 107, 111, 119-122, 125-126, 128, 132, 161, 177, 235, 243, 250, 252, 273, 285, 287 Knowledge Promotion Reform 211-212 labour adjustments 156, 158-159, 162-164, 168, 173, 175-176, 178 labour market 1, 3-5, 11, 13, 15, 36, 40, 42, 44, 47, 53, 56, 61, 72, 77-78, 80, 101, 106, 122, 127, 136-140, 142-143, 146-147, 150-153, 155, 157, 161, 171, 173, 176-177, 179-180, 183, 186, 188, 191, 195-198, 209, 211-212, 215, 224, 230, 262, 269, 272, 274, 276, 279, 282-285, 287, 291 layering 1-2, 9, 12-13, 20, 37-38, 42, 44-45, 48, 50-51, 55, 57, 64, 78, 93, 127, 212, 214, 222, 224, 232, 236, 238, 245, 250, 258, 261, 265, 267-270, 273, 275, 286 liberalization 2, 16, 34, 55, 89, 104, 105, 113-114, 117, 153, 156, 200, 204, 215, 217, 282, 292 local governments 259, 270 macro-corporatism 29 manufacturing 36, 43, 57-59, 65-67, 71-72, 141, 156, 158, 161, 163-166, 168-169,172-176, 178-179, 181, 203-209, 213-215 multi-party systems 264 national insurance 223, 225, 229, 232 needs assessment 248, 251-252 negative feedback 70-73 negotiation 5, 7, 10-12, 17-18, 22-28, 31, 34, 63-64, 73, 75, 84, 127, 161, 175 negotiation myopia 18, 22, 24-25, 31, 63-64, 73 New Public Management 7, 9, 12, 21, 242, 257, 260 normative theory 118-119, 131 Norway 2-6, 8-9, 11-12, 16-17, 21, 36-41, 43-44, 48, 54-56, 59-60, 66, 76-77, 79, 81-82, 85-86, 88, 92, 95-98, 100-101, 103-104, 106, 111-116, 118, 123-126, 130, 132, 136, 139-142, 146-147, 158, 161-164, 169, 174-177, 179-182, 184-185, 189-190, 196, 199, 201-209, 213, 215-217, 220, 227-228, 235-236, 239-240, 242, 247, 249, 259,

 297

261-262, 264, 266-268, 270-271, 276, 278-279, 282, 285, 288-290, 292 organized labour 85, 150-151, 179, 196, 215, 221, 225, 227, 229, 236, 239, 285 parental choice 12, 266, 280 party competition 262-264, 275, 280 path dependency 7-8, 58, 63, 96-98, 222, 231, 285, 289 penalty defaults 10, 18, 26, 28-31 pension 4-5, 11-13, 20, 30, 38, 145, 163, 220, 222, 225-236, 238-240, 284-285, 289-290 pension reform 13, 220, 225-227, 229-233, 235-236, 238-240, 289-290 policy agency 97-98, 109 positive feedback 8, 56-58, 64, 70-73, 265 power 3-6, 9, 15, 22, 26, 28, 33, 36-38, 40, 43, 45-47, 50-56, 59-60, 62-64, 66-68, 71, 74-75, 77-78, 80, 83, 90-92, 105, 115-117, 122, 125, 133, 149-150, 153, 155, 157, 159-161, 164-166, 169, 172-178, 182-184, 190, 196-197, 199, 202-203, 214-216, 221, 227, 229-230, 233, 238, 240-242, 244-246, 251, 255, 259-260, 263-266, 273-274, 276, 279, 282-284, 286-288, 290, 292 power mobilization theory 227 private service sector 179-180, 214 Production Councils 64-65, 70, 91 productivity 4, 17, 44, 48, 51, 61, 65, 67, 69-70, 88-90, 92-93, 120-122, 124, 126-130, 147, 157, 161, 166-168, 171 professionals 243, 246, 254-255, 258-260 property rights 1-2, 11, 15, 30, 36-37, 39-41, 43, 46, 50-54, 59-61, 63, 69, 73-74, 77, 79-80, 82, 89-92, 115, 287-288 Public Limited Liability Companies 49, 99 purchaser-provider model 250, 252, 253, 255-256, 258 recognition 41, 121-123, 126, 128, 132, 141, 191, 205 redistribution 36, 107, 122, 130, 132 redundancies 162-163 reform 1, 3, 5-6, 9-14, 20, 22, 25, 33-34, 37-38, 40, 42-44, 46-51, 53, 58, 68, 77-79, 83, 86-88, 93-96, 108, 110, 113, 115-117, 132, 155, 201-203, 207, 209-215, 220-240, 242, 246-248, 251, 253, 255, 257-261, 263, 265, 269, 271-272, 274-276, 278-279, 284-286, 288-292 Reform 94 202-203, 209-211, 213-215

298 

 Index

regulations 4-7, 22, 26, 39, 46, 51, 65, 67-69, 78, 80, 98, 105, 107-108, 112-113, 121-122, 124, 126, 128-130, 136-140, 142-147, 149-151, 153-154, 157-163, 166-168, 175-176, 179-184, 186-190, 192, 195-198, 201, 204, 207, 214, 224, 234, 248, 291 resilience 56-58, 62-63, 68, 71, 73, 221, 239, 244, 248, 289 retrenchment 5, 220-221, 223-224, 226-228, 232-233, 235, 238-239 shareholder 46, 78, 85, 87-89, 92-94, 99, 108 short-time working 139, 163 skills certification 207 Skytøen Committee 47, 50, 53, 79, 90-91 social contract 161, 171, 174, 175, 176 social partners 4-5, 22-23, 25-26, 28-30, 56, 61, 63, 66-70, 77-78, 80-81, 83, 88, 91-92, 104-105, 124-125, 127, 136-139, 141, 144, 148-149, 151, 153, 157, 179, 181, 189-190, 192, 196, 198-199, 201, 203, 206, 208, 210, 212-214, 224-226, 234, 236, 238-239, 287 stability 6-7, 9-10, 13, 16, 20, 29, 31, 36, 39-40, 44, 58, 97, 126, 136, 138, 153-154, 159, 174, 176, 217, 222, 227, 244, 264-265, 270, 274, 276, 278, 286, 289, 292 stakeholder 43 Sweden 3-6, 12, 16, 21, 27, 33, 36-37, 44, 56, 76, 95, 116, 140-142, 153, 158, 161-163, 169, 172, 174-175, 177-178, 201, 215, 241, 250, 260, 262, 264, 267-268, 270, 273, 276, 278-279, 287, 290 technical expertise 27-31

temporary layoffs 9, 13, 136-148, 151-153, 155, 158, 163, 165-166, 169, 172-174, 285 trade union 3-4, 8, 11, 36, 39-42, 44-45, 80, 83-84, 90-91, 93, 150-153, 155, 158, 178, 180-181, 183-186, 191-192, 196, 198, 221, 249, 255, 282, 285, 287, 290, 292 training offices 209 transparency 89, 185, 192, 244, 248-249, 252, 258 tripartite cooperation 4, 136, 140, 148-150, 153, 202-203, 208, 213-214, 227-228 tripartite negotiations 179, 196, 224 unemployment benefits 13, 17, 136, 140, 142-145, 147-148, 151-152, 163, 285, 290 value competition 266 veto points 8, 57-58, 64, 71, 222, 236, 238, 239 vocational education and training (VET) 12, 201-202 , 204, 206, 210-211, 213-214, 216 vocational schools 38, 205-208 wage bargaining 4, 7, 11, 125, 155, 161, 191, 284, 287, 290 welfare state 1-2, 4-5, 11, 15, 17, 33-34, 40, 54, 56-57, 74, 101, 115, 133, 219-222, 236, 238-240, 261, 263-264, 275, 278-279, 284-285, 287-292 women in management 106, 110, 112 Worker Committee 90 working conditions 4, 22, 40, 59, 62, 74, 77, 92-93, 158, 179-184, 187-188, 190-193, 195-196, 198, 249, 252, 288, 291 Working Environment Act 37, 42, 77, 83-84, 92, 180, 184, 190 zero-sum bargaining 22