Developing Institutions. Collective Agreements in the Dutch IT Industry

Developing Institutions Collective Agreements in the Dutch IT Industry

Developing Institutions Collective Agreements in the Dutch IT Industry ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. mr. P.F. van der Heijden ten overstaan van een door het college voor promoties ingestelde commissie, in het openbaar te verdedigen in de Aula der Universiteit op woensdag 17 mei 2006 te 10.00 uur door Adrianus Andreas Godefridus van Liempt Geboren te Waalwijk

Promotiecommissie: Promotores: prof. mr. P.F. van der Heijden prof. dr. A.C.J.M. Wilthagen Overige leden: prof. dr. W. Buitelaar prof. mr. E. Verhulp prof. dr. K.G. Tijdens prof. dr. P.L.M. Leisink dr. R.H. van het Kaar Faculteit der Rechtsgeleerdheid Cover design: Adriaan van Liempt ISBN 10: 907383880 0 ISBN 13: 978907383880 2 2006 by Adriaan van Liempt All rights reserved. Save exceptions stated by the law, no part of this publication may be reproduced, stored in a retrieval system of any nature, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, included a complete or partial transcription, without the prior written permission of the author, application for which should be addressed to author.

For Linda

Contents LIST OF TABLES AND FIGURES LIST OF ABBREVIATIONS ACKNOWLEDGEMENTS III III III 1 INTRODUCTION 3 1.1 Problem Definition 3 1.2 Institutions and Institutional Theory 3 1.3 Origin and Change of Institutions 3 1.4 Explaining the Emergence of Collective Agreements in the Dutch IT Industry 3 1.4.1 Hypothesis 1: The Limited Choice Set Hypothesis 3 1.4.2 Hypothesis 2: The Efficiency Hypothesis 3 1.4.3 Hypothesis 3: The Anachronism Thesis 3 1.5 Methodology and Outline of this Study 3 2 THE HISTORICAL AND LEGAL CONTEXT OF THE DUTCH COLLECTIVE AGREEMENT 3 2.1 The Historical Context of the Dutch Collective Agreement 3 2.1.1 Worker Coalitions: From Prohibition via Toleration to Legalization (c.1400 - c.1870) 3 2.1.2 Development of Worker and Employer Coalitions (c.1870 - c.1918) 3 2.1.3 The Road to Legislation of the Collective Agreement (c.1900 - c.1918) 3 2.1.4 The Inter Bellum Period (c.1918 - c.1939) 3 2.1.5 Collective Agreements as a Means to an End: Wage Control (c.1940 - c.1970) 3 2.1.6 The Retreating Government (c.1970 - c.2004) 3 2.1.7 Conclusion 3 2.2 The Legal Context of the Dutch Collective Agreement 3 2.2.1 International Regulations and Treaties 3 2.2.2 Government Regulations and Legislation 3 2.2.3 The Employment Contract 3

II 2.2.4 Collective Agreements 3 2.2.5 Extension of a Collective Agreement Through the WAVV 3 2.2.6 Works Council Agreements 3 2.2.7 Conclusion 3 3 HISTORY AND DEFINITION OF THE DUTCH IT INDUSTRY 3 3.1 The development of the Dutch IT industry 3 3.1.1 The Roots of Information Technology 3 3.1.2 The Development of the IT Industry Between 1970 and 1990 3 3.1.3 Specialized IT Training and Education During the 1960s - 1980s 3 3.1.4 The IT Industry During the 1990s 3 3.1.5 The Rise of the Dutch IT Industry in the Late 1990s 3 3.1.6 The Post-Millennium Crisis of the Dutch IT Industry 3 3.1.7 The Three Lineages of Software and Service Companies in the Dutch IT Industry 3 3.2 A Definition of the IT Industry 3 3.2.1 The Many Branches of the IT Industry 3 3.2.2 Defining the IT Industry 3 3.2.3 The Dutch IT Industry in Figures 3 3.3 Conclusion 3 4 INDUSTRIAL RELATIONS IN THE DUTCH IT INDUSTRY 3 4.1 Employers Associations and Branch Organizations in the Dutch IT Industry 3 4.1.1 AWVN 3 4.1.2 Branch and Employers Associations Until 2001 3 4.1.3 Nederland-ICT 3 4.1.4 Werkgeversvereniging ICT 3 4.2 Unions and Works Councils in the Dutch IT Industry 3 4.2.1 The Big Sleep: Union Activities in the Dutch IT Industry Until the 1990s 3 4.2.2 A Wake Up Call for the Unions: The Crash of the Dutch IT Industry at the Beginning of the 1990s 3 4.2.3 The Heydays of the Unions in the Dutch IT Industry: 1996-1999 3 4.2.4 The Development of Union Strategies Since the 1990s 3 4.2.5 2001-2004, A New Crisis in the Dutch IT Industry: Hello Works Council, Goodbye Unions? 3 4.2.6 Summary 3 4.3 The ICK Collective Agreement 3 4.3.1 Origins of the ICK Collective Agreement 3 4.3.2 The First ICK Collective Agreement: Conflicts, Compromises, and Disappointments 3

4.3.3 Extending the ICK Collective Agreement: A Mission Impossible? 3 4.3.4 Collective Agreements Connected to the ICK Collective Agreement 3 4.3.5 A Closer Look at the ICK Collective Agreement 3 4.4 Conclusion 3 5 CASE STUDIES 3 5.1 Introduction and Methodological Account 3 5.2 Case Studies of Companies Using a Company Level Collective Agreement 3 5.2.1 Getronics 3 5.2.2 EDS 3 5.2.3 Atos Origin 3 5.2.4 PinkRoccade 3 5.2.5 Torex-Hiscom (isoft since 2004) 3 5.3 Case Studies of Companies Not Using a Collective Agreement 3 5.3.1 LogicaCMG 3 5.3.2 Ordina 3 5.3.3 Capgemini 3 5.3.4 Centric 3 5.3.5 Baan (SSA Global Since 2003) 3 5.4 Flexibility: Collective Agreements vs. Works Council Agreements 3 5.5 Conclusion 3 6 A SURVEY AMONGST IT PROFESSIONALS 3 6.1 Representativeness of the 2003 Berenschot-Automatisering Gids Salary Survey 3 6.1.1 The Problems With Internet Surveys 3 6.1.2 General Characteristics of IT Professionals 3 6.2 IT Professionals in the Internet Survey 3 6.2.1 IT Professionals Inside and Outside the IT Services Industry 3 6.2.2 Job Level and Highest Attained Level of Education 3 6.3 Union Membership Amongst IT Professionals 3 6.3.1 Union Membership in the IT Industry Compared to National Figures 3 6.3.2 Union Membership, Age, Income, and Job Level 3 6.4 The Terms and Conditions of Employment of IT Professionals 3 6.4.1 Company Size and the Type of Agreement 3 6.4.2 Type of Agreement Versus Terms and Conditions of Employment 3 6.4.3 Preferences of IT Professionals in the IT Services Industry 3 6.5 Conclusion 3 III

IV 7 THE PRESENCE AND FUTURE OF COLLECTIVE AGREEMENTS 3 7.1 Explaining the Presence of Collective Agreements in the Dutch IT Industry 3 7.1.1 The Limited Choice Set Hypothesis 3 7.1.2 The Efficiency Hypothesis 3 7.1.3 The Anachronism Hypothesis 3 7.1.4 Linking the Empirical Results Back to Institutional Theory 3 7.2 Explaining the (inter)actions of Actors 3 7.2.1 Unions and Works Councils in the Dutch IT Industry From an Employees Perspective 3 7.2.2 Unions and Works Councils in the Dutch IT Industry: Employers Tools? 3 7.3 Individuality and Flexibility in Collective Agreements: A Paradox? 3 7.4 The Future of Collective Agreements in the Dutch IT Industry. 3 7.5 Some Final Remarks 3 SAMENVATTING (SUMMARY IN DUTCH) 3 LIST OF INTERVIEWED PEOPLE 3 REFERENCES 3 INDEX 3

List of Tables and Figures Table 1.1 Research questions 3 Table 2.1 Clauses and commitments of parties concluding collective agreements 3 Table 2.2 The legal structure of the various terms and conditions of employment available in the Netherlands 3 Table 3.1 Growth of the Internet between 1969 and 2003 3 Table 3.2 Vacancies in the Dutch IT services (SIC 93 class 72) 3 Table 3.3 Services provided in the Dutch IT industry 3 Table 3.4 Combined services branches in the Dutch IT industry 3 Table 3.5 The Dutch IT industry in terms of the Dutch Standard Industrial Classification (SIC 93) 3 Table 3.6 Added SIC 93 subclass 3 Table 3.7 Omitted SIC 93 classes 3 Table 3.8 Number of IT companies in the Dutch IT manufacturing (30) and services (72) industries 3 Table 3.9 Demographics of the Dutch IT industry in terms of company size 3 Table 3.10 Trading results IT services (SIC 93 class 72) 1996-1999 3 Table 3.11 Growth in the number of IT companies compared to the growth in the number of jobs in the Dutch IT industry between 1993 and 2002 (Index 1998 = 100) 3 Table 3.12 Growth of the number of employees in the Dutch IT industry between 1993 and 2002 (SIC 93 class 72) (Index 1998 = 100) 3 Figure 4.1 Three periods of branch and employers associations in the Dutch IT industry 3 Table 4.1 Relative size of unions and union density in the Dutch IT industry 3 Figure 4.2 Actors and the type of agreements in the field of industrial relations in the Dutch IT industry 3 Table 4.2 Types of regulations in the ICK collective agreement 3 Table 4.3 The à la carte terms and conditions of employment 3 Table 4.4 The pay system and job rating system compared 3 Table 4.5 Working hours 3 Table 5.1 Overview of the cases 3 Table 5.2 Interests of the negotiating parties with regard to a company level collective agreement 3

VI Table 5.3 The flexibility regulations contained within the EDS collective agreement and Ordina works council agreement 3 Table 5.4 The flexibility regulations contained within the Atos Origin collective agreement and Capgemini works council agreement 3 Table 5.5 The flexibility regulations contained within the Torex-Hiscom collective agreement and Baan works council agreement 3 Table 5.6 The flexibility regulations contained within the Getronics collective agreement and LogicaCMG works council agreement 3 Table 5.7 The flexibility regulations contained within the PinkRoccade collective agreement and Centric works council agreement 3 Table 6.1 General characteristics of Internet survey respondents compared to other statistical sources (figures are percentages of the total respondents) 3 Table 6.2 General characteristics of Internet survey respondents working inside and outside the IT services industry 3 Table 6.3 Job level versus highest attained level of education and average age 3 Table 6.4 Distribution of main functions 3 Table 6.5 Unions and union membership of active respondents aged 15-64 3 Table 6.6 Union membership amongst IT professionals inside and outside the IT services industry in percentages of the total respondents 3 Table 6.7 Distribution of types of agreement amongst employees in the Netherlands 3 Table 6.8 Company size and type of agreement in the IT services industry 3 Table 6.9 Type of agreement and company size of IT professionals in the IT services industry 3 Table 6.10 Characteristics of IT professionals working with individual agreements in the IT services industry by company size 3 Table 6.11 Type of agreement of IT professionals versus primary and secondary terms and conditions of employment 3 Table 6.12 Preferences of IT professionals regarding the terms and conditions of employment in the Dutch IT services industry 3 Table 6.13 Preferences and opinions of IT professionals in the IT services industry regarding the type of agreement 3 Table 7.1 Original research questions 3 Table 7.2 The empirical data and empirical explanations for the presences of collective agreements versus the hypotheses 3

List of Abbreviations ASP Application service provider ATW Arbeidstijdenwet (Working Hours Act of 1995) AVV Algemeen verbindend en onverbindend verklaring (extension of collective agreements) AWGB Algemene wet gelijke behandeling (the equal treatment Act) AWV Algemene Werkgeversvereniging (General employers association) AWVN Algemene Werkgeversvereniging VNO-NCW (General employers association VNO-NCW) BBA Buitengewoon Besluit Arbeidsverhoudingen (The Extraordinary Decree on Labour Relations) CA Collective agreement CAO Collectieve arbeidsovereenkomst (collective agreement) CBS Centraal Bureau voor Statistiek (Statistics Netherlands) CNV Christelijk Nationaal Vakverbond (National Federation of Christian Trade Unions) CRM Customer relationship management EAI Enterprise application integration ERP Enterprise resource planning ESC European Social Charter FTU Full time unit ISIC International Standard Industrial Classification ITOP IT Ondernemingsraden Platform (Platform for IT Works Councils) FENIT Federatie Nederlandse IT-industrie (Federation of Dutch IT firms) FHI Federatie van Technologiebranches (Federation of Technology Branches) FNV Federatie Nederlandse Vakvereniging (Netherlands Trade Union Confederation) HIO Hogere Informatica Opleiding (school for higher computer science) HTML Hypertext Mark-up Language HTTP Hypertext Transfer Protocol ICT Information and communication technology ILO International Labour Organisation IOP Individueel ontwikkelingsplan (Individual development plan) IPO Initial public offering ISP Internet service provider ISV Independent software vendor

VIII IT Information technology OEM Original equipment manufacturers PC Personal computer SER Sociaal Economische Raad (Social and Economic Council) SOI Sectoraal Overleg Ondernemingsraden in de IT-branche (Sector level Consultation for Works Councils in the IT industry) SOC Standard occupation classification system SOP Software Ondernemingsraden Platform (Software Works Council Platform) STAR Stichting van de Arbeid (Labour Foundation) VIFKA Vereniging van Importeurs en Fabrikanten van Kantoormachines (Association of Importers and Manufacturers of Office Machines) VNO-NCW Vereniging van Nederlandse Ondernemingen-Nederlands Christelijk Werkgeversverbond (Confederation of Netherlands Industry and Employers) VAR Value added reseller VHP Vereniging voor Hoger Personeel (Union for Higher Personnel) V-ICTN Vereniging ICT Nederland (Dutch ICT Association) VONIT Vereniging van Onafhankelijke IT-specialisten (Association of Independent IT Specialists) WAVV Wet algemeen verbindend/onverbindend verklaren van CAO s Act on extending and un-extending collective agreements of 1937) WAZ Wet Arbeid en Zorg (labour and care Act of 2001) WCAO Wet op de collectieve arbeidsovereenkomst (Collective Agreement Act of 1927) WLV Wet op Loonvorming (Wage Formation Act of 1969) WMM Wet minimumloon en minimumvakantiebijslag (minimum wage and minimum holiday allowance Act of 1968) WOR Wet op de ondernemingsraden (Works Council Act of 1998) WSOI Werkgroep Sectoraal overleg Ondernemingsraden in de ITindustrie (Study group Sector level Consultation for Works Councils in the IT industry)

Acknowledgements There IT is! It is 1982, I am fourteen years old, and my dad finally arrives with a big box under his arm. I have to control myself to an extent unbearable for a boy that age, but I manage, and let my dad open the box. Out comes lots and lots of stuff amongst which the user manual... Uh oh, I hear myself thinking, and yes, my dad starts carefully sifting through the pages, but stops quite suddenly and gives my mum a wink. IT, a brand new Commodore 64 home computer, is quickly installed, hooked up to a small black and white TV, and turned on. After a few seconds the magic line appears that would forever change my life: READY. I have heaps of friends who think my passion for computers is just plain strange. To them it is a machine that does things. They did not understand that to me computers were instruments I can control. Something unimaginable in real life at that time with constricting factors such as parents and school. It seems strange that in those days, and also later when the time arrives to decide what study to take and what to become, I never considered the Information Technology industry. I wanted to become a movie director instead and tell stories to people. Obviously, the most sensible path to pursue a career in the movies was to study sociology, learn about society, and have something to talk about. During my studies I became infected with theories about how societies probably worked and even attempted to do my own bit of theorizing. During my studies I met two wonderful people with an even greater passion for theory and society: Frans Kerstholt and Jan van Wezel. They took me under their wings and introduced me to a range of theories from various disciplines and kept me in blissful ignorance of the real life research topics of a social scientist trying to make a living. After having worked at Tilburg University for a year, I started working for the Hugo Sinzheimer Institute (HSI), a brilliant and dynamic research institute for labour and law, associated with the University of Amsterdam. During my first years at the HSI, I researched various fields such as labour, training and education, most notably together with Harm van Lieshout, who gave me a lot of trust and taught me a great many things during the time we worked together. After some time, I wanted to master a field of my own, which became industrial relations in the Dutch Information Technology industry.

X It was predominantly through the efforts and faith of Paul van de Heijden in the research topic that working on this thesis was possible. I would also like to express my warm thanks to a lot of other people important to me during this study. First and foremost the superpowers Ton Wilthagen and Robbert van het Kaar, who were my direct supervisors throughout the whole project. In hindsight, and after having supervised a number of students myself since, my respect for them and the difficulty of their job has increased significantly. Also Martijn van Velzen and his seemingly bottomless pit of altruism, expertise and humor was of great value to the successful completion of this study. Other former HSI colleagues, in particular Frank Tros, Dlanor Reztleb, and Jan Popma, should be thanked for their time and comments. I am also grateful to Astrid Ornstein for making working life at the HSI a living heaven. The expertise of people such as Harrie Gooskens and Gerard Alberts have proven invaluable to me for writing a pioneering chapter on the Dutch history of computing. Something which simply did not exist at the time I met them. Both Bas Linders and Berenschot should be thanked for allowing me to include a number of research questions on their Internet survey. Finally I would like to express my heartfelt thanks for your continuing support throughout the years to all those close to me and the closest of them all, my wife and love, Linda.

1 Introduction During 1997 and 1998 three collective agreements emerged in the Dutch Information Technology (IT) industry. 1 They seemed to come from nowhere, because collective agreements, especially in the Dutch IT software and services industry, 2 had been a rarity until then. 3 Taking into consideration the fact that collective agreements have been around in the Netherlands since 1927 and that in 2003 85 percent of all employees in the Netherlands fell under the scope of a collective agreement (SZW 2004), 4 the relative absence of collective agreements in the Dutch IT industry is quite conspicuous. Obviously this has something to do with the age of the sector, which is, compared to other industries, quite young. This, however, still does not explain why collective agreements have never become as popular in the Dutch IT industry as they have been in other industries. It also raises the question as to why a number of them suddenly did appear between 1997 and 1998. These basic questions underlying this study gain importance if one also takes the following facts into consideration: The IT industry is generally associated with highly skilled and highly trained employees who are expected to be more than capable of negotiating their own terms and conditions of employment; The legitimacy of collective agreements has been under discussion even before they were introduced in the Netherlands in 1927. Collective agreements are sometimes criticized as unwieldy and outdated institutions; Because the Dutch IT industry is a relatively young sector, it does not share the same historical background and traditions regarding industrial relations which have come to be associated with other industries such as the metal industry or the construction industry. So the basic question of this study is: Why are there collective agreements in the Dutch IT industry? 1 Information technology, in the Netherlands, is often referred to as Information and Communication Technology (ICT). Throughout this study however, the term Information Technology (IT) will be used. Despite the obvious relations between Information technology and communication technology, they are still different enough in nature to be identified as separate industries. 2 See chapter three for an elaborate definition of the Dutch IT industry. 3 See chapters four and five for a detailed historical discussion of industrial relations in the Dutch IT industry. 4 See chapter two for a more elaborate discussion of the origin of the collective agreement in the Netherlands.

2 Developing Institutions 1.1 Problem Definition This study aims at a detailed and distinguished case study of the developing governance regime for labour relations in general, and collective agreements in particular, in the Dutch IT industry. The first goal of the study is to map the emergence of collective bargaining in the Dutch IT industry (See table 1.1, 1); and the second goal is to understand and explain its emergence (See table 1.1, 4). In order to understand the position of the collective agreement in the Dutch IT industry it is also important to look at the socio-political context of the negotiation process (See table 1.1, 2), as well as analyse of the contents of the collective agreements in comparison to other types of arrangements (See table 1.1, 3). Which actors have stimulated the emergence of collective agreements, and why? Why have some IT firms introduced collective agreements and others not? Is this a matter of union power and/or management preference? Finally, as the emergence of collective agreements in this sector is a relatively new phenomenon, the future developments that actors in this sector expect will be examined (See table 1.1, 5). Will the number of collective agreements continue to grow? And, is firm-level bargaining the end of the line, or is an industry wide collective agreement conceivable at some point in the future? Table 1.1 Research questions 1 To what extent and in which particular ways do collective agreements emerge in the Dutch IT industry? 2 Which actors (firms, unions, employers associations and others) are involved in this process? What are their opinions on (various types of) collective agreements? What role have they played in enabling (or hampering) the emergence of collective agreements in this industry? What are the reasons for their preferences and actions? 3 What are the contents of these collective agreements in terms of flexibility and individuality? How do the stipulations in these agreements compare to prevailing arrangements on similar topics in similar IT firms that lack a collective agreement? 4 How can the extent and nature of these collective agreements in the Dutch IT industry be accounted for? 5 What are the prospects for collective agreements in the Dutch IT industry? This chapter will start by discussing the theoretical starting point of this study, which is based on the institutional theories provided by Douglass North and Jack Knight. North is an economist and uses institutions to explain the differences in economic performance between countries. Knight is a political scientist and uses institutions to

Introduction 3 explain social conflict, which he understands as distributional conflicts over resources. Sections 1.2 and 1.3 discuss what institutions are, how they come about and how they change. Collective agreements, works council agreements, and individual employment contracts - the three major ways to regulate the terms and conditions of employment in the Netherlands - are all considered institutions in this study. This is so because they regulate the relation between two parties, the employer and the employees, and describe what both parties can expect from each other. This chapter then continues to formulate three hypotheses, which are based on insights from sections 1.2 until 1.4, and may offer explanations for the presence of collective agreements in the Dutch IT industry. Having or not having a collective agreement, at least in the Dutch IT industry, is a matter of choice. Finally, this chapter concludes with a discussion of the methodologies used in this study. 1.2 Institutions and Institutional Theory The goal of this chapter is to formulate a number of hypotheses which offer theory embedded explanations for the presence or absence of collective agreements in the Dutch IT industry. The theoretical starting point will be institutional theory as defined by North and Knight. The basic assumption underlying this study is that actors (i.e. Dutch IT firms) make rational choices within a given institutional setting that restricts their decisions. 5 Before continuing to discuss whether the collective agreement or its alternatives can be regarded as institutions, it is first necessary to shed some light on institutions, institutional theory, and what it is able to explain. At this point it is important to understand that within traditional sociology, institutions have been identified with group survival and group complexity. Within economics, institutions have come to be associated with economic growth and efficiency, or the absence thereof. In this study institutions are defined as: shared concepts between humans in situations of repeated interaction that are prescriptions for behaviour that are enforced by external agency (rules) or by the individuals themselves (norms), or derive from strategies that individuals adopt within the accepted rules, norms and their expectations about how everybody else will behave (Kay 2003, 3). According to contemporary institutional theory institutions are embedded in an institutional matrix. The institutional matrix consists of all institutional systems. An institutional system itself is a system of functionally interrelated institutions (Bush 1987, 1076). 6 In order to bring systematization into the multitude of institutions, institutions are often categorized into three different levels: the constitutional level, the collective choice or 5 The concept of rationality will be explained later on. 6 In chapter two the institutional system, in which the collective agreement resides, will be discussed in greater detail.

4 Developing Institutions policy decision level, and the operational level of individual actions. 7 Because of the interrelatedness of institutions, there is no strict hierarchical division between the institutional levels. Institutions are imperative for the social structuring of our lives; at least that is what sociologists like us to believe, such as for instance Durkheim, 8 who even went as far as defining sociology as the science of institutions. 9 In sociology institutions are commonly defined as webs of interrelated norms and rules (both formal and informal), governing social relationships. This so-called governing of social relationships implies that institutions can be considered as solutions (though on occasion far from optimal or efficient) to recurrent social problems. Institutions thus structure human behaviour. Saying that institutions are solutions though, immediately puts institutional theory into a functional context. It is, however, not a functional statement in the traditional sense, because the function of institutions is not ascribed afterwards by a theorist, but beforehand by those who created the institutions. Institutions are a product of intentional design. Institutions are created for dealing with recurrent social problems. Recurrent social problems take place in situations where people are living together and where problems with coordinating the interactions between these people emerge. If interactions are repeated and the coordination problems of these interactions are not effectively dealt with by means of rules or agreements, these problems will recur. If people are unable to form such solutions when they are essential for the groups functioning and cohesion, these groups will find it very hard to live together or produce those goods that enable the society to continue. An important factor in the discussion of why institutions perform such a significant social function is group size. As groups grow larger the probability of social problems tends to occur more frequently as there are more potential and complex interactions in larger groups. 10 This latter point has also been noted in economics by some authors, notably North. North defines institutions in two ways. In North (1991a, 97), North defines institutions as the humanly devised constraints that structure political, economic and social interaction. In North (1991b, 3), North continues that institutions are the incentive structure of an economy and, therefore, fundamentally influence individual choices. These humanly devised constraints consist of formal and self imposed informal rules. A noteworthy observation made by North is that these constraints reduce uncertainty and create order. In the end institutions define the choice set and although North restricts this 7 With the introduction of international laws and regulations often influencing the constitutional level of individual states, it would make sense to introduce a fourth institutional level of international legislations. 8 Contemporary sociologists also value institutional analysis highly. See for example Brinton and Nee (1998) and Smelser and Swedberg (1994). Both volumes are characterized by their multidisciplinary approach, which incorporates both sociology and economics. 9 Emile Durkheim (1895) Règles de la méthode sociologique. 10 More on this when Mancur Olson s theory of collective action is discussed below.

Introduction 5 set to economics by combining this choice set with the constraints coming from economic laws, this choice set can be defined as the set of contingent social actions. The concept of incentive structures is especially interesting as it relates institutions and institutional systems to economic outcomes. North (1991b, 8) states that economic growth is dependent on institutions that are able to lower the cost of transactions. 11 Incentive structures are the reflected opportunities provided by the institutional matrix. North (1994) gives an example of what he means by these reflected opportunities: if the institutional framework rewards piracy then piratical organizations will come into existence; and if the institutional framework rewards productive activities then organizations firms will come into existence to engage in productive activities. In this study the collective agreement is regarded as an institution that is embedded in an institutional matrix defining the Dutch system of industrial relations. Figure 2.1 in chapter two gives a schematic presentation of the institutional matrix for the Netherlands and illustrates how actors, institutions, and legislation are related in this system. It would go too far to discuss the entire figure at this point as the Dutch system of industrial relations will be discussed in the next chapter, but it is at this point important to understand the following: Collective agreements and Works council agreements can be a part of individual agreements. Collective agreements are not contracts of employment, but rather a set of agreements on the terms and conditions of employment made by an employer or a body of employers with one or more unions. These agreements are imposed upon each individual contract of employment covered by the collective agreement; Collective agreements can only be imposed upon employers through extension of a collective agreement to an industry or industrial branch by the Minister for Social Affairs and Employment. When a collective agreement is extended to an industry, it is called an industry wide collective agreement; Employers organizations and unions can also create an industry wide collective agreement without it being extended by the Minister for Social Affairs and Employment. These types of collective agreements cannot be imposed upon employers and are only accepted voluntarily; Like collective agreements, also works council agreements are not contracts of employment, but rather a set of agreements on the terms and conditions of employment made by an employer and its works council. In the Netherlands a works council's degree of involvement is limited by law. It is up to the employer to extend the level of participation of its works council beyond this minimum level of participation. 11 Transaction costs are costs involved by bringing about a transaction, which do not include the costs of that which is bought or sold itself. Transaction costs can take many forms, such as e.g. the costs of looking for a buyer or seller, et cetera.