2014
Martijn Krupers
10495460, University of Amsterdam
NORMANDIE 87, 3524 RG UTRECHT
Supervisor DR. ROSA SANCHEZ SALGADO Second reader DR. KATJA BIEDENKOPF
NON-STATE ACTORS IN THE EUROPEAN
SOCIAL FUND: EUROPEANIZATION
MEDIATED IN THE CASE OF THE
NETHERLANDS?
i ACKNOWLEDGEMENTS
The writing of this master thesis was far from negligible, considering it was the most fun and challenging project I have experienced during my (long) course of education as a former MBO student. I am sure that the smirk on my face at the time of this writing would not be possible without gratefully acknowledging the expertise and guidance of my supervisor Dr. Rosa Sanchez Salgado, University of Amsterdam. In this respect, I would also like to acknowledge and thank the second reader of this master thesis, Dr. Katja Biedenkopf for taking her time and effort to read through the thesis as well as all the professionals who participated in the interviews. Thank you for helping me collect some very interesting data.
Furthermore, I would like to thank my family, friends and others who supported me throughout the process of accepting that my student life will soon be over and a very big thanks to Linda Huurman and Lindsay Cameron Lipsay for their contributions to earlier drafts of this thesis. Lastly, a special thanks to Monica Lanz, who supported me on all levels.
ii ABSTRACT
In the Europeanization literature, EU instruments in general have often been used to discuss the impact of EU policy on domestic and local levels of governance, including public, semi-public and private bodies. Whilst in earlier academic literature, the focus was mainly on the top-down processes of Europeanization. With the allocation of the partnership principle to European Social Fund (ESF), I discuss how non-state actors are 'Europeanized' through the use of the ESF. Therefore, the focus will be on the role of the non-state actors in the implementation of the ESF in the Netherlands. To understand the concept of Europeanization properly, this study proposes a complementary use of three new-institutionalism theories, that allows for connecting ‘a set of theoretical ideas and hypotheses concerning the interactions between institutional characteristics and political agency, performance and change’ (March and Olsen 2006: 4). I have found that the processes of Europeanization have been mediated by an interactive and complex domestic setting in the Netherlands and non-state actors became, as Radaelli (2002) calls it, ‘retrenched’ (Radaelli, 2002).
iii TABLE OF CONTENTS
Introduction ... 1
1. Defining Concepts ... 9
1.1. Europeanization ... 9
1.2. Civil society and non-state actors ...10
1.2.1. What exactly are the O&O funds? ...13
2. Theoretical Framework: A new-institutionalist perspective ...15
2.1. Justifying the use of multiple theories in New-Institutionalism ...16
2.2. New-Institutionalism ...19
2.2.1. Mechanism one 'Requirements' ...19
2.3. Rational-Choice Institutionalism ...21
2.3.1. Mechanism two 'Usages' ...22
2.4. Sociological-Institutionalism ...23
2.4.1. Mechanism three 'Social learning' ...24
2.5. Research Design ...25
2.5.1. Selection of method: Process tracing ...27
2.5.2. Data collection and justification ...29
2.5.3. Technique(s) of data analysis ...32
Part II: An analysis of the empirical data 3. New-institutionalism: Requirements ...34
3.1. Principal and Practical Requirements: EU and domestic pressures ...35
3.1.1. The preponderance of sound management over other principles ...40
3.2. Effects of EU requirements in different Actions ...43
3.2.1 Differences between Actions...43
3.3. The power of the Agency: tempering Europeanization with a devilish dilemma ...45
iv
4. Rational Choice-Institutionalism: Usages ...48
4.1. Strategies for pursuing usage in the ESF ...49
4.2. Public value usage: Not just an intermediary role for semi-public bodies ...51
4.3. For-profit usage by private bodies ...57
4.4. A building tension between public value and the for-profit usages ...59
4.5. Change of strategy for tempering the role of private bodies? ...61
4.5.1. Active management strategies ...62
4.5.2. Passive strategies ...69
4.6. The same effects in Action A as in Action D? ...72
4.7. Conclusion ...75
5. Sociological-Institutionalism: Social learning ...77
5.1. The domestic ESF institutional settings ...78
5.2. The shift of domestic agenda-setting: The adopting of EU themes ...79
5.2.1 Differences between the old & new O&O funds ...81
5.3. Skewed opportunities for learning? An analysis of public and private bodies ...83
5.4. Cumulative learning processes eradicated by the dynamics of the principal requirements? ...85
5.5. Conclusion ...86
6. Conclusion: ...87
Literature...92
Appendix ...98
Appendix 1: Interview guide ...98
v LIST OF ABBREVIATIONS
Agency Agentschap Sociale Zaken & Werkgelegenheid
CSO Civil Society Organization
EES European Employment Strategy
ESF European Social Fund
EU European Union
HPA guide Project Administration Manual ESF 2007-2013
MS Member States of the European Union
NGO Non-Governmental Organization
NSRF National Strategic Reference Framework 2007-2013
NSA Non State Actor
NI New-Institutionalism
OMC Open Method of Coordination
OP Operational Program 2007-2013
RCI Rational Choice Institutionalism
SI Sociological Institutionalism
vi LIST OF TABLES
Table 1: Overview theoretical framework; main argument, theory, explanatory mechanisms. ...18
Table 2: Overview of research design, techniques and specifications ...27
Table 3: Interviewees Action A + D ...31
Table 4: Effects of ESF requirements depends on level of fit and possible outcomes ...35
Table 5: Actors categorized on usage in the ESF...48
Table 6: Different strategies for seizing opportunities; differentiated by period-stages and costs ...50
Table 7: Active and passive strategies for semi-public bodies...61
1
INTRODUCTION
What is Europeanization? Does it have to do with European integration or convergence across the European Union, or neither? How does it affect the Member States of the European Union and its governments, the state and society? As Kevin Featherstone (2003) puts it more discretely, ‘how is it evident in the ideas, interests, behavior, and settings of domestic policies? How and why do differences remain?’ (Featherstone, 2003: 3). Moreover, in what way do non-state actors contribute to Europeanization processes? These questions are becoming increasingly important with regard to the growing complexity of the European Union today (e.g. new modes of governance) as well as new problematic issues, such as new social risks1, unemployment, social exclusion and (gender) inequality (Zeitlin, 2010).
According to Simon Bulmer and Claudio Radaelli (2004), there are four macro-dynamics2 that reignited the scholarly debate toward Europeanization3, of which the political argument - an emergent pattern of regulatory competition in the European Union (EU) - is the most interesting. This argument is built on a neo-functionalist perspective, which means that there was a de facto loss in the EU due to competition (a bigger market with a growing economic interdependence between Member States) and a decrease in the visibility of coordination at both EU and domestic levels (less steering capacity) (Zeitlin, 2008, in: Rhodes, 2005: 298; Börzel & Risse, 2003). Meanwhile, European integration kept increasing. By its soft consensus-seeking character, the hope was that the Open Method of Coordination (OMC) would reduce this
1
Examples of new social risks are, as defined by Giuliano Bonoli (2005), the following: ‘Reconciling work and family life, single parenthood, having a frail relative, possessing low or obsolete skills, and insufficient social security coverage’ (Bonoli, 2005: 433-435). An example of an 'old' social risk is the loss of income of the male breadwinner (Bonoli, 2005: 432).
2 In essence, macro-dynamics are the movement in key macro variables for Europeanization (Bulmer & Radaelli,
2004).
3
The other three macro-dynamics, as defined by Bulmer and Radaelli (2004), are: 1. ‘The institutionalization of the single market by the Single European Act (1986); 2. The advent of Economic and Monetary Union, which created not only a single currency and interest rate regime across participant member states (the Euro zone) but also developed an increased interdependency amongst policies; and 3. Process of enlargement, by transferring rules and norms of democratic behavior to new Member States’ (Bulmer & Radaelli, 2004: 1-3).
2
political conflict between Member States (MS), but instead it seems that the strategy became 'part of a dynamic of redefining the social dimension of the EU integration according to the successes and failures of the previous phases and the changing environment' (Pochet, 2005: 72-73). Yet regardless of added political weight4, the problem-solving logic of the OMC strategies was too 'soft' and too ambiguous, being based on deliberation, mutual learning, and good practices, which lacked coordination (Zeitlin, 2010). This was unexpected and disappointing for the European Commission (Commission), because they launched the OMC against the background of a broader governance architecture, the Lisbon Treaty5, to align MS in the domain of social policy to exert its influence in areas where the EU only possesses shared or no competencies (Gürocak, 2011: 1; Armstrong, 2012). Therefore, the Commission decided to substantially reform the method within the second Lisbon Treaty6 in 2006 and the OMC on Social Inclusion and Social Protection (Social OMC) made a shift ‘from social development in its own right to social development as trickle down7 from economic growth’ (Daly, 2010: 149). Despite of the disappointing results and critical modifications, the Social OMC provided Non-State Actors (NSA), such as private organizations, with new opportunities (and constraints) after a marginalized role in the nineties because of economic integration (Pochet, 2005: 72; Börzel & Risse, 2003). Participation of NSAs is increasingly being recognized by the Commission as important for, e.g., transferring EU policy upon the domestic agenda. At first, such organizations were mostly seen as mediators or ‘catalysts’ (Verschragen, et al., 2011) for reducing the political conflict between and within MS by legitimating the EU policy. Today, NSAs participation is
4
The European Employment Strategy (EES), launched at the European Council on Employment in November 1997 and enforced through the Treaty of Amsterdam by the European Commission, was partly inspired by new public management and ‘soft’ governance practices of international (private) institutions (Tulmets, 2010: 300; Rhodes, 2005). This was in essence the foundation for the EU’s development of the OMC (For a detailed historical account see Marlier, 2007, Ch. 2).
5 The Lisbon Treaty, which ‘aimed at making the EU by 2010 “the most dynamic and competitive knowledge-based
economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion”’ (Zeitlin, 2010: 254).
6
For a detailed account see Daly, 2010, p. 147-150.
7
Trickle down in this case means the belief that economic growth will automatically reduce poverty. Consequently, making the social dimension an afterthought or an ‘add-on’ (Daly, 2012: 275).
3
substantially reinforced by EU governance architectures, such as the Lisbon Treaty and the recently launched 'Europe 2020 Strategy'8, by promoting cooperation between NSAs and other key stakeholders within the Social OMC as well as launching new flagship initiatives, such as the Platform Against Poverty and Social Exclusion9 (European Commission, 2010; Brandsen et al., 2005, in: Sanchez Salgado, 2007: 259). However, while the role of NSAs clearly did get reinforced by the Europe 2020 Strategy, the actual participation of NSAs in the Social OMC – for example, being involved in the preparation of the National Reform Plans - was still very limited (Barcevičius et al., 2013: 24). Hence, EU instruments with multilevel access opportunities10 are seen by the Commission as a way to compensate for this limitation, such as the European Social Fund (ESF):
‘The Commission issued a new investment package in February 2013 in which 20 per cent of all ESF funding would be appointed to promoting social inclusion and combating poverty. The access for this funding has been made conditional on the preparation of national poverty reduction strategies involving non-state actors and providing a sufficient evidence base to monitor developments’ (European Commission, 2013, in: Barcevičius et al., 2013: 24-25)
Indeed, the decision of the Commission to align the partnership principle with the allocation of ESF funding may have substantially reinforced the role of NSAs. This decision shifted the central focus of the ESF from ‘how Europe 2020 targets are taken into account by ESF national programming documents’ towards ‘how these targets are translated into concrete actions at the local level’ (Sanchez Salgado, 2013: 1). New opportunities are channeled through the ESF by
8
The follow-up strategy of the Lisbon Treaty launched in 2010 is the Europe 2020 Strategy, which has set up quantified targets in five areas: “employment, spending on research and development and innovation, climate change and energy use, early school leaving and participation in tertiary education, and poverty and social inclusion”. Clearly the focus of the Europe 2020 Strategy is on fighting poverty by making it one of its main initiatives (Daly, 2012: 275; Zeitlin, 2010). Here, I focus on poverty, employment and social exclusion.
9
These flagships initiatives, as defined by Mary Daly (2012: 275-276), are: “engines for boosting growth and are in effect thematic foci for the elaboration of more specific policy issues and tools intended to support the
achievement of the five EU-level targets”
10 The multi-level structure of EU governance is defined as follows: ‘the combination of supranational and
inter-governmental elements; the strength of the judiciary; the functional and technocratic style; the heterogeneity and fluidity of the actors involved over the different policy phases’ (Featherstone & Radaelli, 2003: 9)
4
giving NSAs a European dimension as well as getting a bigger role in the different phases of the implementation stage of the ESF. However, while MS do have an obligation to create a partnership with domestic NSAs, the procedures for doing so are left to domestic authorities (Sanchez Salgado, 2014: Ch. 7). This means that the involvement of NSAs strongly depend on the form and quality of the procedures according to domestic policies of its MS (ibid.). Moreover, the priorities of the ESF are designed at the EU level, while the implementation of it is on the domestic level (Bache, 2007, in: Van Gerven et al., 2014: 2). This makes the ESF not only an EU template, but also an instrument with a complex and heterogeneous governance structure. It shows the importance of studying the role of NSAs in the domestic institutional context to see what kind of resources are made available and what kind of decision-making opportunities there are.
Therefore, these issues lead to important questions, such as 'What is the role and the importance of NSAs at the local level in relation to other levels of governance in the Netherlands? Does a relationship exist between NSAs and higher governance levels, and if so, what does it look like? Moreover, in what way do NSAs strategically use the ESF funds; do they use it to cross-over their own ideas or to promote their cause?' Straightforward answers to these kinds of questions are not clear yet.
The problem is that there is little known about the NSAs’ involvement in the ESF in the Netherlands (Sanchez Salgado, 2014: Ch. 7). Most Europeanization literature focused namely on the impact of Europeanization on domestic policy (cf. Bulmer & Radaelli, 2004, Graziano, 2011), or on the impact of specific new modes of governance, such as the impact of the ESF in MS (cf. Verschragen et al., 2011; van Gerven et al., 2014). However, none of these studies really focused on NSAs involvement in these new modes of governance. There are only a few studies concerned with the role of NSAs (cf. Brandsen et al., 2005; Kröger, 2007, in: Sanchez Salgado, 2014) or social movements (Della Porta & Caiani, 2007.) within the EU. These studies were nevertheless all focused on understanding the involvement of NSAs in the implementation stage of the OMC. It is rather surprising that there is still no research about the involvement of NSAs in the ESF. In addition, it is striking how many Europeanization studies treated the EU only as a
5
political construction, and hence, the analysis of its deeper interactions between institutions and policies remains largely ungrounded and distanced from analyzing the EU as an economic and societal process (Favell & Guiraudon, 2009: 552). Therefore, this study will bridge the literature gap between (historical-) institutional analysis and political sociology by analyzing the involvement of NSAs in the implementation of the ESF as a process (Sanchez Salgado, 2014: 20) with the following research question:
How do non-state actors get involved in the definition and implementation of the European Social Fund in the Netherlands?11
The aim of this question is explanatory as well as descriptive, by pursuing to fill up the literature gap and gaining new knowledge about how these NSAs are 'transformed' through the use of the ESF. An additional advantage of this aim is the allowance for refining existing theory. The purpose of this study is to give an in-depth understanding of the way(s) the ESF funds are being used by NSAs, emphasizing on the implementation of the governance structure with regard to actions carried out and relations of the non-state actors involved.
Sub questions
To what degrees and in which ways has (inter)national governance institutions supported the involvement of non-state actors in the ESF?
How do (inter)national governance requirements for ESF funding affect NSAs?
To what extend do NSAs use the ESF funds into political practice with the intention of pursuing their own political targets?
In which ways does the different institutional settings in the ESF promote the transmission of ideas between NSAs and (inter)national governance institutions?
11
There is an important remark about a key word within this question, namely ‘involved’. Here, ‘involved’ means to what extent NSAs are transformed through means of the ESF. This implies a complex interaction context and the challenge is ‘to identify the intervening causal process – the causal chain and causal mechanism – between the independent variable(s) and the outcome of the dependent variable’ (Bennet & George, 2005: 206-207, in: Checkel, 2005: 9). For a detailed account see ‘research design’.
6 Is there a difference in sociological learning processes provided by economic incentives
and learning processes by interaction opportunities in the ESF?
Studying the role of NSAs in the ESF (and thus within the broader EU governance framework) may be very relevant for society and the Commission itself. NSAs are non-state actors and stand thus close to society. I therefore argue that this makes their role rather important for legitimating European policies in their MS. Amongst the Dutch citizens there is a big gap with regard to a unified European demos or, in other words, an European identity (Della Porta & Caiani, 2009). A Dutch citizen feels Dutch, but in general they do not feel European. Indeed, missing such an identity may consequently end up in problematic issues in terms of legitimating EU policies (Sanchez Salgado, 2014), certainly when the impact of such policies is not clear. This seems to be the case with the ESF in the Netherlands where critical political questions arise about the impact of the ESF on employment in the Netherlands. A random search on Parlis12 resulted into 46 hits alone and therefore one could argue that politically justifying the use of this ESF in the Netherlands without involving some form of civil society would be hard. Especially when looking at the emergence of ‘Euro skepticism’ that exemplifies the growing European identity gap all around Europe13. In addition, all the Dutch citizens pay their taxes and therefore also contribute to the existence of the so-called O&O funds, which, as we will later see, are fulfilling a very important role in the ESF (see also chapter 1). Hence, the studying and understanding of the involvement of NSAs in the ESF is not only social relevant, but important for civil societies and the public governance institutions as well.
Why the Netherlands?
The top priority of the ESF since 2004 is to straighten the inequalities between 'old' and 'new' MS by strengthening the market of new MS. This would enable new MS “lagging behind” to
12
Parlis is a search engine for all Parliamentary papers in the Netherlands. Derived on 14th of June, 2014, from:
http://parlis.nl/zoeken/trefwoord::europees-sociaal-fonds/document_soort::kamervraag
13 As all the Euro skeptic parties across Europe managed to book solid victories with the very recent Euro
parliamentary elections in May of 2014, where, in the case of the Netherlands, the ‘Partij Voor de Vrijheid’ (PVV) had remained its four seats.
7
catch up, instead of creating a “race to the bottom” in the EU (Caune et al., 2011: 22). The Netherlands is an old MS and its society is well-developed with a strong economy14. According to the Commission's website, 'the level of ESF funding and the types of projects that are funded differ from one country to another depending on relative wealth'15. So the Commission made three funding categories based on the GDP per head compared to the EU average (EU with 27 Member States). The Netherlands falls in the lowest funding category with a GDP per head that is over 90 per cent of the EU-27 average. Therefore, the impact of the ESF in the Netherlands should be rather limited in comparison to new MS, where the ESF funds are significantly larger. This makes the Netherlands a 'less likely' case (Flyvbjerg, 2006), because I am expecting less impact of Europeanization through the ESF in the Netherlands for the simple reason that they receive less ESF funds. In other words, possible effects of Europeanization to be found in the Netherlands is representative for Europeanization processes observable in other MS and so, investigating the effects of Europeanization in the Netherlands improves the generalization of this study.
Outline of this study
The key concepts in this study will be defined in the first chapter from an extensive literature study. The second chapter gives a detailed account of the analytical framework focused on new-institutionalism, where I will use three different mechanisms of new-institutionalism for deriving testable hypotheses. The focus of this chapter is also on giving a complete overview of the research design and how the data will be collected. In the second part of this study, a careful and throughout analysis of the collected empirical data will formulate an answer to the derived hypotheses from the theoretical framework. In total, there are three chapters of data analysis. Starting with chapter three, the goodness of fit between NSAs and the ESF requirements will be analyzed. In chapter four, I will analyze to what extend the NSAs use the ESF funds into political
14
all Dutch provinces are among the best 25% of the regions in the EU15. Derived on 13th of June, 2014, from:
http://www.agentschapszw.nl/subsidies/esf-algemeen/operationeel-programma-esf
8
practice with the intention of pursuing their own political targets. The last analysis chapter, chapter five, shows the different institutional setting which provides for different learning opportunities coming from economic incentives as well as interaction opportunities. I will conclude this study with chapter six, in which an answer is formulated to the main research question and the sub-questions of this study. Here, the three mechanisms will be placed in perspective, thereby elaborating on the implications of this study.
9
1. DEFINING CONCEPTS
1.1. EUROPEANIZATIONTo understand important changes occurring in our politics and society, Europeanization is a useful entry-point (Featherstone, 2003: 3). However, as Goetz (2001) has argued, there must be a certain caution in using Europeanization as a cause in search for an effect (Goetz, 2001: 211, in: Bulmer, 2007). Europeanization is not itself a theory, but rather a phenomenon which we are trying to explain by theoretical underpinnings (ibid.). This is why it is essential to define Europeanization and link it to relevant theories (see also chapter 2). Still, defining the term is not so easy. There is an ongoing scholarly debate over what the term actually means16 (cf. Featherstone & Radaelli, 2003).
Basically, the concept is a ‘victim of its own success’ (Sanchez Salgado, 2014considering that Europeanization has a lot of dimensions and definitions. There is no common definition. Nevertheless, Claudio Radaelli (2003) offers a useful definition for the purpose of this study, where rationalist and sociological institutionalisms are incorporated together with a historical-institutionalism perspective:
‘Europeanization consists of processes of a) construction, b) diffusion and c) institutionalisation of formal and informal rules, procedures, policy paradigms, styles, ‘ways of doing things’, and shared beliefs and norms which are first defined and consolidated in the EU policy process and then incorporated in the logic of domestic (national and sub national) discourse, political structures and public policies.’ (Featherstone & Radaelli, 2003: 17)
I will adopt Radaelli’s definition for my research study, because it incorporates the effects of Europeanization to the institutional structures, processes, procedures of government as well as incorporating patterns of norms, values and identities that together comprise the complex interactive relations between NSAs and the EU/domestic governance. Indeed, this
10
institutionalist leaning’ definition of Europeanization allows me to conduct a full and thorough analysis of domestic changes in terms of the ‘policies, practices and preferences’ outlined in the theoretical framework (Adshead, 2013). More precisely, I will focus on the role of actors while identifying the mechanisms that are the interactive link between the domestic and the EU level (Featherstone, 2003: 13).
Another advantage of this definition is that it has been arguably proven to be difficult to measure the impact of Europeanization in macro-economic terms, while the impact is well-measured at the micro level of governance (cf. Vandenbroucke & Vanhercke, 2014: 71; van Gerven et al., 2014; Verschraegen et al, 2011, Sanchez Salgado, 2013; 2014). It is somewhat surprising that most Europeanization studies defined Europeanization as ‘building European capacity’, whereby it is best explained by theories of European integration (Bulmer, 2007). Defining Europeanization in this way is related to the domestic impact, also called by Börzel (2002: 193, in: Bulmer, 2007: 48) a top-down dimension. However, I am after a finer-grained analysis by understanding Europeanization from a bottom-up perspective and, thus, treating Europeanization upon domestic agenda setting not as a cause - what changed? – but rather as a process - how and why did it change? - (Bulmer 2007: 49; Sanchez Salgado, 2014). Europeanization is thus defined as a transformation process, where NSAs are confronted with European and/or domestic requirements and norms and beliefs about the ESF.
1.2. CIVIL SOCIETY AND NON-STATE ACTORS
Theory-oriented political scientists have complained that the concept civil society is ‘vague’, considering that most civil society theorists offer ‘widely varying current usages, interpretations and perspectives on the term than something more rigorous’ (Heinrich, 2005: 212; Jensen 2006: 39). This clearly indicates that the concept of civil society is diverse and multifaceted. Therefore, as argued by Armstrong (2002), there is no such thing as a unified civil society where all decisions are collectively made with a single set of targets derived from a common public (Armstrong, 2002: 107). Henceforth, the preponderant view of civil society organizations (CSO) in today's Europeanization literature is thus understood as 'organizations in society that
11
distinguish themselves from the state, market and usually the family where collective action and other forms of engagement takes place' (Heinrich, 2005: 212; Deakin, 2001: 7; Elliott, 2003: 7; Sanchez Salgado, 2014: 56). Or, more star, as voluntary non-profit organizations (Salamon et al., 1999). This is, in part, thanks to the Commission, which enforced the role of ‘social partners’ due to several occurrences in the EU at the international and European levels, such as the legitimacy crisis of the European institutions and in particular that of the Commission. Hence, the definition of social partners in the White Paper (2001) has been made broader than in previous papers that just emphasized Non-Governmental Organizations (NGO) (Simsmans, 2003: 479)17. In the 2001 White Paper, social partners are defined as:
‘the sum of all organizational structures whose members have objectives and responsibilities that are of general interest and who also act as mediators between the public authorities and citizens’ (Smismans, 2003: 482).
This definition is an institutional expression of civil society, because the Commission switched its focus from representation of ‘the European citizen’ to representation of ‘the organized citizen’, namely the intermediary organizations comprising civil society’ (Ibid.: 482). Therefore, the emphasis lies with the social interaction between economy, citizen and state, and civil society is defined as a mediator between them (Warleigh, 2002). With it the latest scope of social partners on the Commission’s website does not only include NGOs, but also ‘trade unions, employers, professional associations (organizations representing social and economic players that are not social partners in the strict sense of the term), charities, grass-roots organizations, and religious communities’18.
However, the concept of CSOs is not so useful in this present study, considering the strong neo-corporatist system in the Netherlands. In such a neo-neo-corporatist system, it is not uncommon for private bodies starting to collaborate in order to represent the interests of their sector in formal
17
See also Armstrong (2002) for an excellent discussion.
12
decision-making (Van de Bovenkamp et al., 2010). In other words, some private bodies may receive key tasks of governance ‘through corporatist arrangements or norm production through collective bargaining’ (Armstrong, 2002: 108). This is exactly what happened with private bodies in the case of the ESF. According to the 'Agentschap Sociale Zaken & Werkgelegenheid' (hereinafter: the Agency), the 'Opleidings & Ontwikkelingsfondsen' (hereinafter: O&O funds) are private bodies that are controlled by the social partners and aim to optimize the sector labor markets (also known as sector funds or schooling funds). Each sector in the Dutch labor market is represented by an O&O fund and the Agency made the O&O funds the formal applicator of the ESF and thereby receiving tasks of governance in the ESF (see the section below for further elaboration on this).
The problem is that the concept of CSOs does not work well in a neo-corporatist system like the Netherlands, since there are bodies that are both private organizations as well as mediators with a public function as mentioned above. Hence, the Dutch domestic institutional setting differs significantly from other MS, such as Spain, where a centralized strong government alone took upon this public function and not non-state actors (Sanchez Salgado, 2013). Therefore, I will not use the concept of CSOs and simply make a distinction between public, semi-public and private bodies. Here, the semi-public and private organizations are both non-state actors in the sense that they are actors who naturally exist beyond the state (Simsmans, 2003). Notice that semi-public bodies are still private organizations, but are carrying a public function by receiving key tasks of governance. One may argue that I should define these actors as public bodies, considering they have a public function. However, defining concepts with regard to just their function has rather normative implications (Sanchez Salgado, 2014),
To conclude, the public bodies are defined as all governance institutions, such as the Agency or municipalities. The semi-public bodies are non-state actors that carry a public function. For example, the O&O funds that want to improve and protect the general interest of their sector. Private bodies on the other hand is defined as non-state actors that do not carry any public function and pursue a for-profit usage in the ESF, such as companies or consultancy bureaus. In essence, every private body pursues two broad financial targets which are: 1. Maximizing their
13
profitability and; 2. finding the strongest financial structure possible to ensure continuity and growth. The essence of the for-profit usage is deduced from these targets. Note that this agenda-setting is significantly different from the agenda of NGOs or similar organizations and therefore I am excluding such non-profit organizations from this definition19.
1.2.1. WHAT EXACTLY ARE THE O&O FUNDS?
This is an extra section concerning the semi-public bodies, the O&O funds. The role of these funds in the ESF is rather hard to understand for laymen and therefore, it is necessary to clearly state what they are and what purpose they pursue.
The O&O funds are private organizations that are controlled by the social partners and aim to optimize the sector labor markets. Each sector in the Dutch labor market is represented by an O&O fund. If the O&O fund has a direct relationship to a collective labor agreement (CAO)20, then the social partners (trade unions and employers' organizations) could ask the Minister of Social Affairs to make the private fund generally binding. This generally binding declaration imposes obligations on employers and their employees (who are not connected with the relevant industry associations) such as ceding a certain percentage of the wage bill per employee. This means that all companies and workers in the sector are to make a financial contribution to issues that benefit the entire industry. This prevents mere private bodies of free-riding. Therefore, one could conclude the O&O funds are not pursuing a for-profit usage, but rather a public value usage. According to the CAO, this percentage will go to the fund so that they can provide funding for every individual employee for training purposes (wholly or partly).
19 I have observed little to no participation of NGOs in the ESF themes (also known as Actions) which were analyzed.
Therewith I am not saying that NGOs do not participate at all in the ESF, but including these organizations in this present study is not relevant, as I have encountered only for-profit organizations. That is why I use the term 'private bodies' instead of NSAs or NGOs. The focus is on the agenda-setting of these organizations.
20 This is a set of agreements between employers and employees. These agreements are bargained with employers'
organizations and trade unions together. A CAO contains agreements on issues, such as working hours, compensation, child care, education, vacation, pension and severance arrangements.
14
For each training and/or project the client must submit an application to the O&O funds. The amounts of the funds are sector-specific.
The O&O funds were thus created for the purpose of improving and protecting the general interest of their sector. Considering they already took on a subsidy role in their sector, it was rather logical for the Agency to make them also the formal applicator of the ESF, as was regulated in the OP21. Hence, they received the ability to provide private institutions ESF funding for an extra impulse to their training courses in Action D22. Their role in the administration of the ESF became thus very specific, since they took the same public function and responsibilities as the municipalities in the ESF. This meant that the private bodies, such as sector companies, could only apply to these O&O funds. This intermediary idea was enforced by the fact that the Agency reserved a certain amount for each sector per round in Action D23. So each sector received a certain percentage of the total ESF budget from the Agency annually24. This is elaborated further in chapter 4.
21 This regulation only applies for Action D. For Action A the municipalities are the formal applicator. See also Paper
26.642Europees Sociaal Fonds (ESF), derived on 13th of May, 2014, from:
https://zoek.officielebekendmakingen.nl/kst-26642-121.html
22
See also: https://www.otib.nl/subsidie/branche_breed_opleiden_2014/
23 OP, analyzed by author.
24 Each year in the ESF period 2007-2013, there was a new round where you could apply for ESF funding as
15
2. THEORETICAL FRAMEWORK: A
NEW-INSTITUTIONALIST PERSPECTIVE
As stated in the Introduction, Europeanization itself is not a theory, but rather a concept that needs strong theoretical underpinning. It is important to use a ‘good’ theoretical framework25 that is applicable to the concept of Europeanization. The theory of new-institutionalism (cf. Pierson, 1996; Börzel, 2005) fits with such a framework and focuses on institutional development, which is considered among the most relevant factors in shaping political life (Sanchez Salgado, 2014). Using such a framework allows me to seek and identify a number of ‘theoretically predicted steps’ (Checkel, 2005: 15) as causal mechanisms of Europeanization. These steps are hereinafter called 'mechanisms' because 'they are largely used instrumentally by the EU' (Grabbe, 2003: 312). The aim of the theoretical framework is bringing the theory closer to ‘what really goes on in the world’ (Checkel, 2005: 1).
In this study institutions are understood to be ‘organized setting within which modern political actors must typically act’ (March and Olsen 2006: 4). This definition allows for connecting ‘a set of theoretical ideas and hypotheses concerning the interactions between institutional characteristics and political agency, performance and change’ (ibid.: 4). For example, the ESF platform that is a working group for recalibrating new ESF regulations, organized by the Agency, is an institution where public and private bodies must interact. Therefore, this may be a good institution for testing the new-institutional hypotheses.
25 A ‘good’ theory must lead to logically derived hypothesis that are falsifiable. To analyze the applicability of a
given hypothesis, it is essential to ‘use a style of negative case testing by intentionally seeking negative or unique cases that do not fit the hypothesized relationship’ (Berg, 2006: 323). Searching such cases leads logically to the conclusion that my theoretically predicted intermediate step(s) is/are false. This is (or even more) as important as verifying a hypothesis. Falsification of hypothesis makes research not only the more interesting, it also helps to prevent the ‘logical fallacy underlying post factum explanations’ (Merton, 1968, in Berg, 2006: 322).
16
It is important to consider that a general theory such as new-institutionalism may be vulnerable to problems of ‘concept stretching’26 by extending such theories to broader contexts (Collier & Mahoney, 1996). In order to reduce this potentially conceptual validity problem, I use different theories in a complementary way. In addition, I use ‘subsumption’27 with the identified mechanisms of new-institutionalism, which also allows uncovering a deeper understanding of the process as well as avoiding the ‘kitchen sink argument’ (see ‘research design’ for a more detailed account).
2.1. JUSTIFYING THE USE OF MULTIPLE THEORIES IN NEW-INSTITUTIONALISM
The framework is based on earlier work of authors (cf. Knill and Lehmkuhl, 1999; Börzel and Risse, 2003), by drawing upon the new-institutionalism arguments. The following arguments will be used in this study:
The 'goodness of fit’ argument: ‘the ease of adaptation to European policies depends upon the extent to which these fit national policies and institutions’ (Mastenbroek & Kaeding, 2006: 331);
The ‘logic of consequentialism’: the effect of institutions on the choices and constraints of the actor.
The ‘logic of appropriateness’: the effect of institutions on the norms and behavior of the actor;
Together these three arguments form the core of the new-institutionalist literature (cf. March & Olsen, 2006; Knill & Lehmkuhl, 1999; Pierson, 1996; Hall, 1993; Verschraegen et al., 2011) by representing three well-known and developed perspectives of new-institutionalist theory,
26 See also Giovanni Sartori (1970), 'Concept Misformation in Comparative Politics," American political Science Re-
view.
27
Here, subsumption is used as the mechanisms together explain the outcome. However, mechanism A derives the other mechanisms B and C. In other words, if mechanism A is removed or replaced by another mechanism, then mechanism B and C explanatory power becomes zero in explaining the outcome. For example, if you apply a project to the ESF then there are certain requirements for applying. If you do not conform to this requirements then you cannot participate in the ESF and learn from other projects or strategies (usage and learning).
17
namely: new-institutionalism itself, rational choice-institutionalism and sociological institutionalism (Featherstone, 2003). These three perspectives are complementary in the sense that they do not exclude one another but tend to focus on different explanatory mechanisms (March & Olsen, 2006). Using theory in this way provides two advantages: First, it avoids academic isolation, which has been pointed out by Featherstone & Radaelli (2003) as the main risk of researching Europeanization (Featherstone & Radaelli, 2003: 20). A second advantage is that it enlarges the scope of my conceptual lens, but the framework should, of course, be seen as a simplification of reality (see also Bulmer & Radaelli, 2004 for an excellent discussion). However, there are also some concerns about the use of new-institutionalist arguments. One concern with regard to the validity of the goodness of fit argument is raised by Mastenbroek & Kaeding (2006). They argue in the wake of Börzel and Risse (2003) if misfitsui generis is either a necessary or a sufficient requirement for Europeanization (Mastenbroek & Kaeding, 2006: 333-334). Nevertheless the topic under study is embedded in an EU model that is in turn embodied in the Europe 2020 Strategy, which makes the ‘goodness of fit’ argument certainly valid (Bulmer & Radaelli, 2004: 9). Dutch actors applying for funding are participating in the ESF and thus, are also participating in the broader Europe 2020 strategy. Hence, they may have a certain fit with the targets set by this EU strategy. For example, a NSA concerned with the environment or animal abuse is not likely to receive ESF funds, since these objectives do not match with those of the Commission. In other words, the ESF may create a significant impact on the shaping of NSAs in the Netherlands by imposing specific requirements supplementary to those of the Commission (Sanchez Salgado, 2007). In such a situation the goodness of fit argument works well. Still one must realize that the goodness of fit argument is embedded in a top-down perspective and represents just one of the mechanisms of Europeanization (Bulmer, 2007). This again underlines the essence of a complementary use of the multiple new-institutionalist perspectives. Indeed, a very careful investigation by synthesizing different perspectives and taking context of multi-level governance into account adds value and importance to the term Europeanization and provides an in-depth understanding of the involvement of NSAs in the definition and implementation of the ESF in the Netherlands (Featherstone, 2003: 19).
18
Also the framework of rational choice-institutionalism (RCI) being used as a theoretical underpinning for Europeanization needs some further elaboration. Applying RCI, i.e. the logic of consequentialism, to the analysis may run a teleological risk. Featherstone (2003) is right in asking if the choices of actors seizing opportunities are really independent in explaining the outcome (Featherstone, 2003: 14). In other words, whatever NSAs do, the outcome becomes the revealed preference (Thelen, 1999: 375). In the length of this argument, Bartels (1996) warns that it is rather easy to make the assumption that the units of analysis have more complete information than they actually do in the real world. Indeed, assuming that NSAs are fully informed is easier than to carefully define and actually demonstrate their use of information (Bartels, 1996). I agree with these critical arguments in being cautious when using RCI. Still as argued before, this approach may be useful for its heuristic purposes, which is getting and understanding the whole picture.
In the following sections of this chapter, each perspective and its explanatory mechanism will be presented separately by using the arguments as described above. I am using one main argument per theory, but it does not mean that the other arguments are not relevant and/or treated in the same theory. The theories as well as the arguments are complementary in this sense. However, using a main argument per theory allows me to clearly identify a valid mechanism as is shown in Table 1.
Table 1: Overview theoretical framework; main argument, theory, explanatory mechanisms.
Main argument Theory Explanatory Mechanism
The goodness of fit argument New-Institutionalism ESF requirements The logic of consequentialism Rational Choice-Institutionalism Leverage effect The logic of appropriateness Sociological Institutionalism Social learning
19
2.2. NEW-INSTITUTIONALISM
New-Institutionalism (NI) is mostly at the level of mid-range theory, meaning its scope involves a limited range of cases that are unified in space and/or time (Thelen, 1999: 373). New-Institutionalists are usually concerned with observed empirical puzzles that exist from comparisons such as the difference in effects of Europeanization with regard to NSAs in advanced industrial democracies (ibid.). Here, the focus is upon how ‘domestic and or EU institutions have an effect on NSAs preferences and interests in the short term’ (Bulmer & Burch, 1998, in: Featherstone, 2003: 7) 28. Looking at the interactions between the actors involved is not only avoiding a overly deterministic approach - i.e. the ‘stickiness’ of path-dependency as a reaction to Europeanization (see also Thelen, 1999) - but also being able to zoom in on the capacity of NSAs to benefit from these opportunities and avoid constraints. Indeed, NI underlines a goodness of fit argument between the EU & (sub-) national institutions and NSAs. So to what extend and in which ways has the EU supported the involvement of NSAs in the ESF? And in what way do the ESF requirements affect NSAs?
2.2.1. MECHANISM ONE 'REQUIREMENTS'
NSAs receive certain benefits and costs by participating in the implementation of the ESF. However, as Heather Grabbe (2003) argues, the adaptation costs are far removed from the eventual reward of accession (Grabbe, 2003: 321). In other words, ESF requirements come in one big step: 'Accession to the funds is bound to an overall readiness, and not tied to partial readiness' (ibid.: 321). So technically speaking, NSAs that seek ESF funding for their projects but differ from the Europe 2020 Strategy targets can thus be rejected, if the project is found not to be ‘ESF proof’ by the public bodies, such as the Agency. For instance, there are certain requirements for being eligible to ESF funding, such as the partnership principle, 'Sound Financial Management ’ and the additionality principle (article 11/14/15 of Regulation (EC), No
28 I will not focus on the long-term, because the Europe 2020 Strategy has been launched very recently in 2010.
Moreover, the decision of the Commission to align ESF funding with the partnership principle is even more recent (Barcevičius et al., 2013: 24-25). That is why the focus is not on the long-term, but rather upon the role of time in the short-term.
20
1083/2006)29. These requirements signify that the projects must contain a close cooperation between the national, regional and local level; if there is a proper management and control systems in order to enhance transparency of the initiated projects through the ESF (Gürocak, 2011); and if the ESF funding does not fully replace public or other form of expenditures. In addition, there is a vast timetable for requesting ESF funding which certainly limits the scope of NSAs for using some kind of stalling techniques. Indeed, the public bodies have more scope to use timing as a trade-off than NSAs, which has to comply with all the requirements before getting access to the funds. Therefore, the following hypothesis is derived:
Hypothesis 1: The more stringent EU requirements for ESF funds are, the less NSAs will be eligible to apply for the ESF funds.
Belonging to the Continental European welfare regime30, the Dutch (sub-) national institutions are embedded in a strong social insurance system with a substantial role for social partners (Palier, 2010). The funds are distributed by the SZW agency, a central authority, to the main beneficiaries such as the O&O funds or the 408 municipalities, which set up and implement different programs and projects in which NSAs are involved. Minna van Gerven et al. (2014) defines the Dutch institutional setting thus as a simple and compound polity by using the terminology of Schmidt (Schmidt, 2006, in: van Gerven et al., 2014: 4), because of the strong consensus-oriented politics (also known as 'poldermodel') and neo-corporatist policy making (ibid.: 4). The Dutch (sub-) national institutions thus have a certain misfit with the EU institutions (see also Börzel & Risse, 2003), which may open up opportunities for NSAs at the local and/or regional level, considering the requirements for ESF funding are based on the quantitative targets set out in the Europe 2020 Strategy and not based on the local/regional level. On top of
29 Derived on 1st of April, 2014, from:
http://ec.europa.eu/regional_policy/sources/docoffic/official/regulation/pdf/2007/general/ce_1083(2006)_en.pdf
30
Structural reforms of the 2000s may significantly changed the Continental Welfare regime leading to a shift away from traditional protection of income and status maintenance towards 'employment and market-friendly welfare systems' (Palier, 2010: page number is not known). This shift is important as significant to be aware of. However, for the sake of simplicity and pragmatic reasons, I will treat the Dutch welfare system as part of the Continental Welfare regime.
21
that, the Agency is obliged to take the EU targets into account within ESF expenses and the additionality principle (article 15 of Regulation (EC), No 1083/2006)31 obliges '(sub-) national governments to spend EU funds in addition to any planned expenditures' (Verschraegen et al., 2011: 65). This may result into a grey area for testing the ‘proper’ programs and or projects adjusted to these requirements before establishing binding rules upon them, which may in turn create more room for NSAs to maneuver and resulting into lower accession costs. Hence, a non-state actor participating in ESF funded projects managed by the municipalities may have more bargaining power (Gürocak, 2011: 9). The assumption here is that NSAs are selected for the project by the municipalities because of their local knowledge or other unique features. This is very different from NSAs applying their own projects for ESF funding. That is why the following hypothesis is derived:
Hypothesis 2: NSAs participating in ESF funded projects initiated by municipalities have different adaptation costs than NSAs initiating their own ESF funded projects in the Netherlands.
2.3. RATIONAL-CHOICE INSTITUTIONALISM
Rational choice institutionalism (RCI) is more concerned with finding counterfactuals in research, than understanding how actors and institutions have been affected (Featherstone, 2003). The focus of RCI is typically on how opportunities are created and/or opened up by institutions and how actors are exploiting these opportunities. For example, if a European public space is shaped by interactions with EU and domestic private institutions than Europeanization does only exist because 'actors perceive that their interests are better represented at the EU level or that they decide to seize European incentive structures' (, 2014: 25-26). Thereby Sanchez Salgado is suggesting that the EU offers opportunities which differ per type of organization. In other words, the EU is affecting the chances of mobilizing and/or realizing its collective interests depending on one’s agenda-setting.
31
Derived on 1st of April, 2014, from:
22
As argued by Koopmans (1999), the concept of opportunities is rather troublesome, considering there is a tendency to overstretch the concept in most Europeanization literature (Koopmans, 1999: 95). He argues that “opportunity" is rarely defined by scholars, but it ‘generally refers to constraints, possibilities, and threats that originate outside the mobilizing group, but affect its chances of mobilizing and/or realizing its collective interests’ (Ibid.: 96). Hence, opportunity refers to the possible options for action, which may be either available or not. Therefore, the concept of opportunity also refers to a notion of uncertain outcomes, or as the author put it: ‘The chance that certain options will bring about desired outcomes and the risk that they will have undesired outcomes’ (Ibid.: 96-97). So I will follow Koopmans’ definition of opportunity, which defines opportunities as: ‘Those options for collective action that are (1) available and (2) expected to result in a favorable outcome’ (Koopmans, 1999: 97). This definition can satisfactorily be applied to the NSAs in the ESF by putting emphasis on the importance of institutional environment in the shaping of organizational behavior (Sanchez-Salgado, 2007) as well as on the agenda-setting of the actors.
2.3.1. MECHANISM TWO 'USAGES'
In the Netherlands, the ESF helped to increase resources available for labor market and social inclusion policies. At the same time, the framework of multilevel governance within the ESF is being modified through the ESF funded projects that involve NSAs (Caune et al., 2011: 224-225).
The question is therefore how much of this change is to be explained by the involvement of NSAs? In what ways do NSAs use the ESF funds into political practice with the intention of pursuing their own political targets (Verschraegen et al., 2011; van Gerven et al., 2014). These targets can range from influencing a particular polity decision to increasing one’s capacity for action or one’s access to the political process. It is thus important to know what the specific targets of the NSAs involved in the ESF are, which determines how the ESF will be used as a leverage effect (see also Zeitlin and Pochet, 2005). In other words, do NSAs seize the available opportunities and if so, how, or are they blocked by institutional or political obstacles - also known as veto-points - (Bulmer, 2007)? Hence, Usage as a mechanism seems justified as interesting, which is defined as 'the transformation of EU resources into political practice with
23 the intention of pursuing a specific goal' (Erhel et al., 2005; Jacquot and Woll, 2003: 6, in: Verschraegen et al., 2011: 56). For example, at times of real retrenchment (e.g. the Euro crisis) the ESF can serve for NSAs as a strategic advantage over rivals through funding and pushing their own agenda at the local level. NSAs may analyze the usage(s) of the ESF through a cost/benefit analysis and look for the best way to maximize their utility. Also, the partnership principle of the ESF (article 11 of Regulation (EC), No 1083/2006) obliges cooperation between all levels of governance in the preparation and implementation of ESF funded projects, making it thus a fruitful ground for NSAs accessing higher governance institutions, such as the national (Agentschap SZW) or EU-level (Commission) (Gürocak, 2011). As a result of the previous notions, the following three hypotheses are derived.
Hypothesis 3: NSAs use the ESF as a leverage effect to gain strategic advantage over rivals Hypothesis 4: Usage of the ESF by NSAs is to increase their own capacity in terms of strengthening their internal organization.
Hypothesis 5: The ESF is used by NSAs as an access opportunity to national and/or European institutions.
2.4. SOCIOLOGICAL-INSTITUTIONALISM
Sociological-institutionalism (SI) is concerned with the realm of norms, ideas, organizational culture and the psychology of politics (Bulmer, 2007). SI mostly underlines the logic of appropriateness and, therefore, focuses on the role of actors in 'practices and political interactions (...) whether they are institutional, ideological, political or organizational.' (Palier et al., 2011: 10; Woll & Jacquot, 2010). In other words, SI conceptualizes ideas as transfers in mechanisms of social learning and explores the transmission of ideas from the EU to the domestic level and vice versa (Pedersen, 2006). For example, the changing discourse in the framework of multilevel governance in the ESF may have spurred new learning processes among NSAs (Caune et al., 2011: 224-225).The focus is here on the relationship between public and semi-public bodies as well as the relationship between semi-public and private bodies,
24
considering that the logic of the argument suggest that local/regional environments are always embedded in larger national or transnational ones (Finnemore, 1996).
The a priori assumption is that the ESF automatically provides interaction opportunities for social learning processes. Yet one may argue that this does not necessarily have to be the case. The dominant discourse of 'soft power’32 in the area of social policy may, according to Radaelli (2008), prevent social learning on the regional and/or local level, considering the absence of any form of 'hard' power - that is EU legislation - (Radaelli, 2008: 251). However, the ESF is embedded in EU legislation and thus has more 'bite' than in comparison to the Social OMC, so investigating the idea of social learning in the ESF seems interesting as well as justified.
2.4.1. MECHANISM THREE 'SOCIAL LEARNING'
The argument is that social learning33 as mechanism leads to institutional change, such as an increasing similarity between organizations (Powell & DiMaggo, 1991). However, learning per se can take many different forms and may be hard to measure. Therefore, it is rather essential to operationalize the mechanism(s) of SI interaction than can valid and reliably measure the concept of learning. These mechanisms are based on principles such as ‘interdependence, frequent exchanges, trust, diplomacy and reciprocity’ (Rhodes, 1999, as cited in Sanchez Salgado, 2014: 231). So when private bodies engage in interaction opportunities provided by the ESF institutional settings, they engage into social learning by, inter alia, frequent exchanges, observing and trust with others actors (Hemerijck & Visser, 2001, in: Radaelli, 2008: 243).
For analytical purposes, I will use the typology of Visser (2009), who builds on the work of Checkel (1998) namely: 'Thin' learning and 'thick' learning (Visser, 2009, in: Van Gerven et al.,
32 According to Radaelli (2008), the Treaty base for European Union policy is thin or non-existent and the 'diverging
political views hinder the development of law' (Radaelli, 2008: 239). Therefore the Social OMC is mainly based on the Council’s guidelines, which is significantly different as ‘hard’ law that is obligatory. Soft power is thus not obligatory, but rather ‘friendly pressing to make the other comply’.
33 I lend the definition of social learning from Hall (1993): 'a deliberate attempt to adjust the targets or techniques
of policy in response to past experience and new information. Learning is indicated when policy changes as the result of such a process' (Hall, 1993: 278).
25
2014: 6). The first mechanism of learning occurs when private bodies simply 'learn how to use' the ESF (Radaelli, 2008: 244). It suggests an interest in the ways in which one’s own ideas, behavior and culture correlates with and is driven by European Union’s cultural factors rather than local governance task demand (Finnemore, 1996). For example, private bodies may have learned on the basis of trust that strategies are more effective in reaching their targets. Hence, the next hypothesis will be:
Hypothesis 6: private bodies want to actively engage into a sociological process of learning in the ESF.
Thick learning on the other hand implies that private bodies become increasingly aware of other paradigms (Van Gerven et al., 2014). As Finnemore (1996) argues, the involvement of private bodies in such EU instruments is culturally necessary, considering this participation ‘constructs or constitutes what organizations want or, in the case of the EU participation, what they are’ (Finnemore, 1996: 338). They may therefore radically redefine their agenda-setting on the basis of evidence coming from other ESF funded projects or programs (Woll & Jacquot, 2010). For example, organizations may change their agenda-setting significantly in order to take part into other ESF themes (also known as Actions) as a consequence of frequent exchanges or participation in the ESF (Radaelli, 2008: 244). Therefore, it produces the following hypothesis: Hypothesis 7: private bodies change their preferences towards a new agenda-setting after entering a sociological process of learning in the ESF.
2.5. RESEARCH DESIGN
The EU is not new, it has already been in place, and it produces decisions and impacts on its MS (Bulmer & Radaelli, 2004: 3). The literature review in the introduction made a convincing case that Europeanization happens, yet as the review indicates; it differs on exactly why and how it happens (Alderson, 2001, in Checkel, 2005). This is mostly because of the tendency to focus only on the outcome and mostly ignoring how the outcome came together as a process. To understand the outcome, I will use process tracing, also known as within-case analysis, in a qualitative manner as a method for identifying the workings of the casual mechanisms of
26
Europeanization in a given context (ibid.). This means analyzing changes at the domestic level and then assessing how the EU has contributed to this change. The main advantage of this research design is that it avoids bias towards EU-level explanations, by controlling other possible outcomes that may account for these changes (Sanchez Salgado, 2014: 42)
As Collier & Mahoney (1996) rightly points out, there is a potential problem in terms of validity concerns using the concept of Europeanization as an outcome, which has namely one value – the impact of Europeanization- (Collier & Mahoney, 1996: 68). This means that the outcome has no variance and entails a no-variance problem34. Consequently, it would be arguably hard to draw direct causal linkages without some comparison techniques. Researching Europeanization as a bottom-up process through the means of the ESF in one country may make the involved NSAs indeed a homogenous set. In addition, I acknowledge also the fact that it would be rather naïve to expect that domestic change is driven by Europeanization only (see also Van Gerven et al., 2014: 4).
So, why keep sticking to the no-variance design? First, the design is very useful for exploring new information and or new explanations (Collier & Mahoney, 1996). Second, a ‘country-by-country’ comparison may arguably not be the best method in the case of the ESF. EU social policy is not an extrapolation or replica of domestic social policy, making the amount of ESF funding per country differ significant. In other words, the EU does not determine the domestic institutional arrangements of its MS (Bulmer, 2007: 53; Barcevičius, et al., 2013). In addition, NSAs organizational capacities and characteristics depend on the available resources that also differ significantly per country (Sanchez Salgado, 2007). For these reasons and other pragmatic ones35, it is arguably better to use a no-variance research design in this study. However, it may be interesting to compare the NSAs in different ESF themes (see also Ch. 4.6). Nevertheless, I
34
No variance on the dependent variable (Collier & Mahoney, 1996). This problem also occurs when looking at a range of variation and treat it as a single outcome, for example 'high' or 'low' growth (ibid.: 72).
35
Pragmatic reasons are i.e. limited time and money resources. A solution is focusing on NSAs in the Netherlands which makes the scope of my research feasible.
27
acknowledge the no-variance design as the main weakness of this study and hence, caution must be used in interpreting the results, considering that the NSAs are operating under the same roof, namely in the ESF in the Netherlands.
Table 2 gives an overview of the research design intended to be used within this study. Process tracing is a good method for such a research design by carefully testing hypotheses against multiple perspectives of Europeanization (Collier & Mahoney, 1996). A vast majority of scholars of methodological writings has suggested that in this way the power of causal inference is greatly enforced (see ibid.: 70). This will be elaborated in more detail below. In addition, strategically selecting cases within this research study allows me to explore the set of mechanisms in the sense of negative/positive case testing, in order to assess whether these mechanisms can plausibly be understood as producing the outcome.
Table 2: Overview of research design, techniques and specifications
Research Design Techniques Specifications
Method: Process tracing “causal steps”
Data collection: Semi-structured interviews Strategic selection of cases
Unit of analysis: Non-state actors In the Netherlands:
Selected by characteristics and organizational capacity of each specific organization.
Data analysis: Content analysis Analytic categories shaped by certain question
asked during the course of the interview process.
Source: Author.
2.5.1. SELECTION OF METHOD : PROCESS TRACING
Process tracing as a method is embedded in post-positivist philosophies and in scientific realism, because of its focus on mechanisms and its causal dynamics that are ‘epistemologically
