Understanding European disintegration

(1)

Understanding European disintegration;

a case study on the

renationalisation of the discretionary competence to prohibit the

cultivation of genetically modified crops.

Master thesis by Finn F.W. Roelofs

For the completion of the International Relations track of Political Science

Nijmegen School of Management

Supervised by Dr. Thomas R. Eimer

14-8-2017

(2)

(3)

(4)

Abstract

This thesis answers the question how we can explain European disintegration. It does so

through the analysis of the highly controversial debate on the cultivation of genetically

modified crops in the European Union. The competence to prohibit such cultivation is

reallocated to the member states and this thesis asks why this has occurred. Other European

policies remain adamant to calls for subsidiarity. Through the analysis of this disintegrative

process, this thesis enhances our understanding of integrative dynamics. By employing the

classical theories of European integration Neofunctionalism and Liberal

Intergovernmentalism to explain disintegration, the explanatory scope of these theories is

increased. As this thesis will show moreover, the Liberal Intergovernmentalist explanation

underestimates the role the European Commission plays, and the Neofunctionalist approach

(5)

Abbreviations

ALDE Alliance of Liberals and Democrats

CAP Common Agricultural Policy

CDU Christlich Demokratische Union Deutschlands

CSU Christlich-Soziale Union

ECJ European Court of Justice

ECSC European Coal and Steel Community

ED European Disintegration

EFDD European of Freedom and Direct Democracy

EFSA European Food and Safety Authority

EI European Integration

EP European Parliament

EU European Union

FOEE Friends of the Earth Europe

GMC Genetically Modified Crop

GMO Genetically Modified Organism

GUE-NGL European United Left – Nordic Green Left

AMIF Asylum Migration and Integration Fund

LI Liberal Intergovernmentalism

MS Member State

NF Neofunctionalism

OLP Ordinary Legislative Procedure

QM

Qualified Majority

SPD Sozialdemokratische Partei Deutschlands

UK United Kingdom

UMP Union pour un Movement Populaire

US United States (of America)

(7)

(8)

Chapter 1: Introduction

In March 2015, decades of debates within the tripartite leadership of the EU consisting of the European Council (the Council), the European Commission (the Commission), and the EP, have resulted in an agreement to let the MSs decide for themselves if they wish to ban the cultivation of GMCs on their territory. Although the integration project has gained increased attention and criticism, which marks the end of the so called ‘permissive consensus’ of continued EI (Hooghe and Marks, 2009), the project has never been put in the reverse gear. The reallocation of GMC authorisation competences is an unprecedented and remarkable disintegrative event in the EU. The

reallocation is remarkable in the sense that it effectuates a breach of one of the corner stones of the EU: the internal market. Although the internal market perhaps exists with some inconsistencies, it’s scope has always expanded rather than reduced. Why have no other European policies been subject to disintegration? The CAP has remained a European competences in spite of high controversy over the years. Moreover, the AMIF which addresses the highly salient topic of migration has not been disintegrated. In contrast, GMC policy has been renationalised; how can this difference exist? The following question arises: ‘’How can we explain the reallocation of discretionary competences

regarding the authorisation of GMCs?’’ This thesis aims at gaining an understanding of the process of ED through the study of a rare instance where EU competences have been reallocated back to the MSs, namely: the discretionary competence to prohibit the cultivation of GMCs. It does so by describing the conditions under which theories of EI, namely Liberal intergovernmentalism (LI) and Neofunctionalism (NF) expect disintegration to happen, and applying them to the case. The objective is to conduct a case study analysis with the aid of both theoretical approaches in order to demonstrate which elements of the explanations can be confirmed by the empirical material, and can explain why a EU competence is reallocated back to the MSs. An important assumption for this thesis is that the process of ED is institutionally driven. The institutions which are considered relevant this reallocation, although each theory assigns more (or less)

(9)

importance to different ones, are the supranational institutions of the EU, most notably the Commission and the EP, societal actors such as Greenpeace, Friends of the Earth Europe (FOEE), or business actors such as Monsanto, Syngenta and Bayer, and the governments of the most powerful MSs of the EU: the United Kingdom (U.K.), Germany, and France. To answer this thesis’ question, two theories of integration will be used: LI and NF. Both theories are renowned for their explanatory power when it comes to integrative dynamics. Their capacity to address a reversal of these developments has however not yet been properly tested. This is what this thesis addresses: the possibility to make use of the well-known theories of integration to explain disintegration.

NF explains integration as a process whereby interdependency between states creates ‘practical’ incentives for integration, which itself turns into a perpetuating process. This process is expected to be advanced and promoted by societal actors with supranational loyalties or interests. This thesis will show that the NF understanding of the core integrative mechanisms can be used to generate expectations for disintegration as well. It is expected that NF explains disintegration through ‘practical’ reasons of protecting initial more important policy objectives, and that this process is advanced and promoted by the same kind of societal actors as those who promote integration. LI expects that integration is the result of sovereign states who cooperate in order to seek absolute gains. State preferences reflect domestic bargaining outcomes which, in turn, can be predicted by the domestic actors’ ability to successfully organize and represent their interests. European integration, in this sense, is a means to ensure the survival of the MS rather that the replacement of it (Hoffmann 1995, p. 89). In order to explain disintegration then, one might be able to turn around the above explanations and generate expectations about disintegrative forces. It is expected that LI explains disintegration as a

consequence of these MSs’ preferences again. These preferences are based on expectations of absolute gains from disintegration. This thesis functions as an assessment of how both LI and NF explanations of EI succeed in explaining the reallocation of EU GMC cultivation policy. By doing so, it increases the scope of LI and NF to explanations of disintegration, rather than

(10)

integration alone. Moreover, it generates evidence about when to expect disintegration in the EU. For the project of EI, the reallocation of GMC authorisation competences may not have drawn such attention as the ‘Brexit’, it is nonetheless a unique and illustrating example of disintegration. Together with the pending British exit from the EU, the reallocation of GMC authorisation

competences are amongst the first examples of ED. Similar to the Brexit, Council Directive 2015/412/EC appeals to public claims to subsidiarity; the principle that calls for the allocation of discretionary competences at the government level that is most appropriate. The reallocation can therefore be understood as part of the broader debate on subsidiarity and the rising momentum of support for a different EU. Although calls for more subsidiarity are not new, little has been achieved in name of this principle until now. When the circumstances of this reallocation are analysed, we might be able to generate new knowledge about when we can expect disintegration to happen.

Addressing these problems, first the theoretical framework will be described; juxtaposing the two theoretical frameworks proposed in the

introduction by highlighting their biggest differences as well as by describing how they account for EI combined. Secondly, the method of research is discussed and hypotheses based on the theoretical chapter will be introduced and operationalised. Third, the empirical data will be analysed and the hypotheses will be evaluated. The final chapter concludes this thesis. Here the main question will be answered and evaluated, the strengths and weaknesses of this thesis will be discussed, and

(11)

Chapter 2: Theoretical Framework

2.1 Neofunctionalism

In order to explain how discretionary competences are reallocated towards a supranational level, the theory of NF stages a political struggle for these competences within the boundaries of the EU and its MSs. Actors or geopolitical concerns that lay outside the EU are excluded in its understanding of integration. It expands the state-centric view of realist scholars in IR by including supranational actors, as well as societal actors such as transnational corporations, employers organisations, environmentalist organisations and other interest groups as decisive in international politics. In the distinction between these internationally operating societal actors and the political actors operating domestically, it is especially pointed out that the actors and developments on the European level pose a tremendous source of demand for further integration. Theoretically speaking then, it is possible to understand why and how EI proceeds by analysing the EI process as a function of these actors’ interests.

‘Process’ is a key term in this regard, as NF understands integration as a self-perpetuating progression that delivers its own incentives for further integration. What is taken here to be

understood as the integration process, is described by one of the first and most prominent scholars of EI Ernst Haas (1958, p. 16) as a process: ‘’whereby political actors in several, distinct national settings are persuaded to shift their loyalties, expectations and political activities towards a new centre, whose institutions possess or demand jurisdiction over the pre-existing national states’’. This process has a political and a social element to it (Niemann and Schmitter, 2009, p. 46-47). In Haas’ definition the social element is highlighted by the ‘shifting of loyalties, expectations and political activities’. The political element is highlighted by the possession of jurisdiction over the pre-existing national states, or: the pooling of discretionary competences. It is especially the latter that is of interest here, as this thesis seeks to explain the reallocation of discretionary competences. What is useful about the distinction between the social and political element to integration however, is that it highlights that

(12)

there is a political struggle for integration that is somehow determined or influenced by the more social or structural conditions of the European playing field and vice-versa: it’s an interaction.

An interaction moreover, at the expense of the power and capacities of the national states. This is different from the idea of federalism which aims at an understanding of the accommodation and preservation of the differences in interests and identities of the parts of a legally constructed whole (Burgess, 2009, p. 26). The theory of NF is however grounded in the functionalist idea that such differences gradually disappear in international political interactions, rather they being accommodated. This idea has become well-known through the works of David Mitrany, who emphasises the functional logic that connects international politics across the world: ‘’[N]o matter how great the historical, constitutional and social variations, similar problems under similar conditions led to similar ways and means for dealing with them …. A rare thing in social experience’’ (Mitrany, 1975, p. 136). So, NF is not only about the shifting and merging of national political activity, it is also about the changing nature of politics. This relates to cosmopolitan ideas and to ‘the end of ideology thesis’ (Bell 1962; see also Steffek, 2015) which state that ideology is no longer relevant in governing. In a world that witnesses increased economic traffic, communication, and other interactions, politics increasingly consists of providing solutions to practical problems, or: ‘’…’the administration of things’ (Rosamond 2000, 58).

This brings us back to the societal actors who are at the foundation of integration. In this sense, NF is different from Historical

Institutionalist (HI) explanations of integration. HI theorizes political integration as the consequence of actors’ previous choices, whereby they become ‘locked in’ and make future choices constrained by previous ones, or even become ‘path-dependent’ (Pollack, 2009, p. 127). In contrast: ‘’In […]

functionalist explanations, political institutions are assumed to have been deliberately designed by contemporary actors for the efficient performance of specific functions […]’’ (Pollack, 2009, p. 127). Schmitter (1969, p. 164) for example emphasises that, in contrast to this rather ‘technocratic’ view of international politics: ‘’ […] the neo-functional logic actually emphasizes the contrary: the role of

(13)

bureaucratic practice, organizational ideology, and the creative intervention of administrative and

political elites’’. An empirical reality for the EU

according to authors of integration research, is that there is increased involvement of all sorts of supra- and subnational actors in the making of policy (Marks et al. 1996, p. 341). This has sparked notions such as ‘policy networks’ and ‘Multi-level Governance’ which emphasise ‘interdependency’ and ‘mutuality’ between different sorts of public actors, and is focussed on studying the policy networks and the links between the actors (Peterson, 2009, p. 105). The NF acknowledges too, that all sorts of actors are important to integration: political parties, workers and employers organisations, transnational companies, supranational actors, think-tanks and advisory organisations, and

bureaucratic elites are examples of such actors. In contrast however, it focusses on actors’ interests and conflict as the main determinants of integration politics. The interplay between these actors thus forms the locus of attention for NF. Politics is a group-based activity wherein multiple actors all try to influence policy. A diverse set of actors operate at the domestic and international level and forge coalitions across countries and organisations (Niemann and Schmitter, 2009, p. 48). It however never loses sight of actors’ self-interested motives. The assumption is that actors act in a self-interested way (Niemann and Schmitter, 2009, p. 48), and that therefore the: ‘’[A]nalysis of political process must give a central place to the phenomena of group conflict’’ (Lindberg, 1963, p. 9). EI in this sense, is deliberately driven by the societal actors who benefit from it: ‘’It starts with egoistic utility-maximizing actors who cooperate to overcome collective action problems’’ (Risse, 2009, p. 147).

To complete the circle, it needs to be said that NF understands that actors’ interests are somehow altered and transformed through the integration process. Interests are derived through a pluralist process where actors learn and alter their

preferences in the context of integration (Haas, 1970, p. 627). This bears resemblance to the social constructivist ideas about how the interaction between actors (or more generally: ‘agency’) and institutions (‘strucure’) is mutually constitutive (Risse, 2009, p. 148). As a consequence: [T]he EU as an emerging polity is expected not just to constrain the range of choices available to, say, nation

(14)

states, but the way in which they define their interests and even their identities’’ (Risse, 2009, p. 148). Indeed, NF emphasises a ‘social element’ to integration, and a ‘shift of loyalties’ towards the supranational level, as mentioned above. According to Risse (2009, p. 147) this implies: ‘‘[S]ome constitutive effects of European integration on the various societal and political actors’’. But although NF scholars such as Ernst Haas do flirt with a more ‘thicker understanding’ of rational choice (Risse, 2009, p. 147), integration according to NF does not follow a ‘logic of appropriateness’ where: ‘’Human actors are imagined to follow rules that associate particular identities to particular

situations, approaching individual opportunities for action by assessing similarities between current identities and choice dilemmas and more general concepts of self and situations’’ (March and Oslen, 1998, p. 951). Instead, NF supposes that the EU provides for circumstances that enable societal actors to engage in positive- rather than zero-sum games (Niemann and Schmitter, 2009, p.48). By these conditions, the institutions are meant which provide for information, transparency and trust (Rosamond, 2000). As a result, the European environment offers: ‘’[A] multitude of different advantages for different groups’’ (Haas, 1958, p. xiii).

Key actors

An important supposition in this thesis, is the focus on agency driven integrative dynamics. This means that the EI project is encouraged by actors who have an interest in integrative schemes. Importantly, societal actors such as transnational businesses or supranational actors themselves are expected to be pro-active as they reap benefits from integration, in contrast with national states which are regarded as reactive (Stone Sweet and Sandholtz. 1997, p. 306), or as Haas (1967, p. xix) puts it: ‘’The economic technician, the planner, the innovating industrialist, and trade unionist advanced the movement not the politician, the scholar, the poet, the writer’’.

Mechanisms

While actors act rationally, NF expects integration to follow the logic of ‘spillover’. According to Niemann and Schmitter (2009, p. 49), the concept of spillover has been applied first in order to: ‘’identify the driving force and inherent logic of integration via increased functional/economic

(15)

interdependence’’. NF understands that spillover pressures are logical or practical motives for actors to pursue integration because of interdependencies: ‘’the process of regional integration (in the contemporary setting) depends on the realization of mutual gains from cooperation in policy arenas characterized by high levels of functional interdependence’’ (Schmitter and Lefkofridi, 2016, p. 3). This is the somewhat uneasy position of early NF: ‘’At some point, the functional logic takes over (spillover) leading to further integration’’ (Risse, 2009, p. 147, italics added). In order to understand integration then, it must be examined when this logic takes over, or when and how it is employed by actors. To meet this problem, three variations of the concept have been developed by Tranholm-Mikkelsen (1991) which enable us to understand the different mechanisms of integration. These are functional spillover, political spillover, and cultivated spillover. These will be explained below. Moreover it will be explained how these mechanisms can be reinterpreted in order to explain disintegration: the reallocation of pooled discretionary competences back to the MSs.

 Functional spillover and spillback

The first mechanism of NF integration is functional spillover. This concept was initially dubbed by Ernst Haas (1958, p. 383), who described an ‘expansive logic of sector integration’ as a process whereby the integration of one sector leads to ‘technical’ pressures pushing states to integrate other sectors’. These technical pressures hold that further integration to other fields of

government is a necessity since the prior field of integration will be undermined if other related sectors do not follow. For example, the decision to remove tariffs on the movement of goods within the EU has illustrated the need for harmonization of regulations which lead to price discrimination; an effect of the removal of tariffs that undermines the initial goal to foster economic cooperation within the EU. These developments will effectuate an increasing

interdependence of MSs. Schmitter and Lefkofridi (2016, p. 3) noted that NF expects this for all MSs to happen: ‘’The distribution of this increase will not accrue primarily to a single Member State (MS) or a set of ‘hegemonic’ MSc within the [trans regional organization]’’. This

(16)

interdependency leads to practicalities that speak in favour of integration, perpetuating the process. Schmitter (1969, p. 164) moreover emphasizes the role of agency for functional spillover: ‘’As transnational exchange rises, so does the societal demand for supranational rules and

organizational capacity to regulate’ (Stone Sweet and Sandholtz, 1997, p. 306).

As functional spillover expects that the basis of the practicality of integration lies with the (economic) interdependencies between the MSs, a logical step for a reinterpretation of this mechanism would be to expect the practicality of disintegration to lay with a lack of

interdependency in a given policy field. However, NF expects interdependencies to increase more and more as the process of integration gets its own ‘impetus’ for continued integration. Still, the concept of spillback was mentioned in 1970 by Lindberg and Scheingold (p. 137) who understood it as ‘’a situation in which there is a withdrawal from a set of specific obligations. Rules are no longer regularly enforced or obeyed. The scope of Community action and its institutional capacities decrease’’. They argue that this spillback can be easily explained by pointing at the original incentives for disintegration. To illustrate this, they explain that the case of spillback in the European coal sector where the problem of a surplus of coal was redefined as a Belgium problem which resulted in its temporary de-facto exclusion of the common coal market. As a consequence there was a reduced willingness to cooperate on the supranational level. To link this to interdependency: it explains spillback as a direct result of negative effects of interdependency as the Belgium surplus can harm others’ interests. Although there is increasing literature on the ‘perverse’ effects of institutionalisation which may undermine social or political support for them (Pollack, 2009, p. 128), these focus on the more ‘structural’ determinants of ‘negative feedback

loops’, rather than on agency. The

mechanism of spillback then can be understood as: the diversifying logic of sector disintegration where practicalities dictate that societal actors’ problems are better addressed through multi-level approaches to the project of EI. Interdependency can present practical reasons, or functional spillback linkages, in two ways. The first one is based on the assumption that the

(17)

project of EI proceeds intermittently such as in the case of the Belgium exclusion from the common coal market. In this case, the integration of a certain policy field is considered to be an obstruction to the achievement of another integrative scheme. The second one is based on the assumption that interdependencies do not limit themselves to the European continent and that it may be more practical in some cases to integrate with other countries at the expanse of the EU.  Political spillover and political spillback

The next form of spillover is called political spillover. It is expected that societal elites such as the representatives of political parties, businesses, and other interest groups, promote integration. They will do so if domestic policies do not promote their interests well enough, and if they expect that their interests are better served at the supranational level. NF expects that these domestic elites undergo a learning process whereby they come to understand that their interests are better served at a supranational level, rather than domestically (Tranholm-Mikkelsen, 1991, p. 5). Increased interaction causes these elites to reorient themselves to the supranational level. Such involvement at the supranational level may also be necessary if they expect their interests to be harmed (Tranholm-Mikkelsen, 1991, p. 4). EI thus generates opportunities and risks that make it rational for domestic elites to refocus their attention to the supranational level. Assuming that it is rational for elites to promote integration, how are they expected to promote this? Haas (1958, p. 312-313) focussed on the pressures which are exerted by non-governmental elites such as leaders of political parties, trade associations and trade unions. Later NF scholars tend to include a wider set of interest groups (Schmitter, 1971, 257). Lindberg (Niemann and Schmitter, 2009, p. 50) on the other hand, attributes greater significance to governmental elites. In order to promote integration, both groups are expected to express the need for an integrative approach. The success of these claims for these elites may depend on the successfulness of bypassing their respective governments. This is more easily achieved when different international elites’ interests align. An example may be that all manufacturers of cars have an interest in integrative measures

(18)

that establish an internal market. Such alignment can result in cooperation through the creation of umbrella organizations. These are focussed on representing the shared interests of domestic elites who seek advantages they cannot achieve domestically and are thus inherently geared towards the promotion of collective solutions. Faced with opportunities and risks at the supranational level, domestic elites thus organize themselves on an international level and are able to lobby more effectively at this level as well. Again, it may be possible to imagine how political spillover might be useful as an explanation for disintegration. The expectation is that

political spillback occurs when interest groups no longer find that integrative schemes serve their

interest. Or at least, that disintegration in certain policy subfields does serve their interest. This is not to say that these groups will retreat from the supranational level; they may have several goals which they pursue that still need this involvement. Once they have established umbrella

organizations and have focussed their activities on a supranational level, it is rational for them to employ this position to exert pressures to disintegrate the supranational organization in certain policy subfields if this better serves their interest; the integrative schemes that serve their interests are consolidated and now they exert pressure on a reallocation of discretionary

competences that serves them best. This sort of two-level is important as Van der Vleuten (2005, p. 485) argues that: ‘’When supranational pressure is not met by domestic pressure (France, 1980s), the government does not go beyond ‘rhetorical implementation’ ’’. Consequently, political spillback pressures are expected to be more effective when umbrella organizations simultaneously exert pressure at the international as well as the domestic level.

 Cultivated spillover and cultivated spillback

A third version of spillover is called cultivated spillover. This type of spillover focuses on top-down rather than bottom-up integrative pressures: those exerted by the important supra-national institutions such as the Commission and the EP. These actors are expected to protect their own interests rather than that of the MSs individually. One way to do this, is by trying to expand their

(19)

mandate. The role played by these institutions is mainly found in mediating the international bargaining process. Haas makes a distinction between interstate bargaining and bargaining facilitated by European institutions such as the ECSC and later the Commission (Niemann & Schmitter, 2009, p. 50). Interstate bargaining, according to him, was characterized by lowest-common denominator outcomes. Instead, results of negotiations facilitated by an

institutionalized mediator tend to be considered as promoting a common interest. The outcome of negotiations, according to Haas, involves a redefining of the conflict (Haas, 1961, p. 367). The relative success of cultivated spillover then, depends on the extent to which the Commission is able to redefine the conflict, or ‘upgrade the common interest’ (Haas, 1964, p. 66). This is not so much done by creating ‘win-win situations’, but rather by making compromises. Common interests, Niemann and Schmitter (2009, p. 50) state: ‘’are upgraded to the extent that each participant feels that, by conceding something, it has gained something else’’. These

compromises then, according to the cultivated spillover logic, are more easily achieved under the auspices of European institutions such as the Commission. The expectation is that the outcome of this redefinition of conflict almost always implies the expansion of the powers of an international agency (Haas, 1961, p. 367). This follows the understanding of supranational actors as self-interested rational actors that seek to increase their power (Moravcsik & Schimmelfennig, 2009, p. 50). The expectation is thus that the Commission (and other supra-national institutions) exerts pressure to expand its mandate as the European Union grows. Lindberg, in addition to Haas, expanded on how the Commission may exert its pressure. He stated that, because of its privileged position of centrality and authority, the Commission can direct the dynamics of interstate relations as well as intrastate relations between interest groups (Niemann & Schmitter, 2009). Within the Commission or other institutions, compromise is more easily achieved as well due to the like-mindedness of officials. These officials may form ‘epistemic communities’ who share a scientific paradigm (Schmitter and Lefkofridi, 2016, p. 4). Consequently their

(20)

later authors described the active role institutions can play in reallocating discretionary

competences (Lindberg 1964, 71; Tranholm-Mikkelsen 1991, 6; Smyrl, 1998). Smyrl (1998, p. 97) for example stated that the Commission can convince MSs to adopt its policy preferences through gradual persuasion and alternative specification of the problem over the course of policy

formulation. It can also influence agenda setting according to Pollack (1997, p. 102), albeit informally: ‘‘Moving to informal agenda setting, supranational institutions enjoy no formal monopoly on the right to set the Council's substantive agenda. Nevertheless, their policy expertise and institutional persistence can provide them with certain informational advantages vis-a-vis both competing agenda setters and the Council of Ministers in a setting of incomplete information’’. In the policy process leading to new proposals regarding the allocation of competences, the Commission, and the Parliament as well, have different opportunities to propose amendments and convince other parties of their point of view and perhaps redefine the problem or its solutions. In this sense the outcome of the international bargaining process is significantly different from a process in the absence of such actors.

Although the supranational organizations within the EU are usually assumed to strive for an increase of their mandate, situations can be conceived of where it is in the interest of these institutions to discharge some of their discretionary competences, and hence support disintegration. This may occur when a certain policy subfield that has been integrated on a supranational scale causes damage to the reputation of a more important integrative scheme or institution. This especially holds for the Commission. Stephenson (2010, p. 1046) argues that in relation to policies carried out, the Commission is never free from scrutiny by MSs, and: ‘’[I]t must demonstrate the effectiveness and efficiency of EU policy in the face of constant pressures to nationalize policy or carry it out at lower levels’’. When a supranational organization perceives a subordinate practice or institution as a threat to its own reputation, it may understand

disintegration as a necessity to its own success or survival. The specific policy subfield might have been politicized and turned into a ‘hot potato’ that the supranational institution might not want

(21)

to burn its fingers on. It may then sacrifice certain integrated policy fields to protect others. However, seeing that the supranational organization is always striving to protect its mandate, the expectation is that it will seek to circumvent handling the issue in other ways as well. This

explanation of political spillback presupposes a hierarchical ordering of EU institutions. Although NF scholars have not acknowledged this explicitly, it is not unlikely that the Commission is more capable of directing policy than some of its executive branches such as the EFSA. This view of cultivated spillback expects that supranational organisations will always be proactive, even when they seek to avoid responsibility they do so by actively striving for disintegration. In contrast, a more reactive or ‘weak’ Commission can undermine its own practices by not acting. Institutions can ‘cultivate the seeds of their own demise’ (Greif and Laitin, 2004, p. 34), increasing pressure for change over time (Pollack, 2009, p. 128).

(22)

2.2 Liberal intergovernmentalism

The LI theory of integration is merited for its usefulness as a foundation for synthesis with other explanations (Moravcsik and Schimmelfennig, 2009, p. 67). It provides for an elaborate and step-by-step approach in understanding international politics, describing how meso-level inferences about integration are derived from actor level assumptions. Similarly to ‘rationalist institutionalism’, LI seeks to address the problem of ‘equilibrium institutions’: ‘’[N]amely, how actors choose or design

institutions to secure mutual gains, and how those institutions change or persist over time’’ (Pollack, 2009, p. 126). It is an application of this approach to the practice of regional integration (Moravcsik and Schimmelfennig, 2009, p. 67).

In contrast with the NF approach to EI, LI does not include a strong social element to

integration but focusses on the legal transfer of competences from the domestic to the international level. This can be associated with the supposed actors deciding on the reallocation of competences, namely: national states’ governments. These actors are legally speaking the ultimate decision makers on their territory and hence decide on a reallocation of competences. States are considered to be ‘Masters of the Treaty’ (Moravcsik and Schimmelfennig, 2009, p. 68), which amounts to the

understanding of legal authority as the main resource of LI’s understanding of power (Marks, Hooghe, and Blank, 1996, p. 9). In this sense, integration is a function of state power and theirs to take back when they want to: ‘’EC politics is the continuation of domestic policies by other means’ (Moravcsik

1991, p. 47). Similar to NF,

LI conceptualises integration within the European context. A central element to the LI explanation of integration is the distinction of politics in two levels: the domestic, and the international (Putnam, 1988). Moravcsik adds a temporal variable to this analysis of international politics through the setting of EI into three consecutive phases: preference formation, the outcome of interstate bargaining, and the formation or adjustment of institutions (Moravcsik and Schimmelfennig, 2009, p. 69). Each of these phases is specified as a theory with its own expectations about its results. During these phases, which will be introduced more elaborately below, actors are expected to act in a rational

(23)

self-interested way. Regional integration according to LI can be understood as resulting from a series of rational, self-interested choices made by governments which are assumed to make a calculation of the utility of their actions (Moravcsik and Schimmelfennig, 2009, p. 48).

LI initially set out to understand the integration process in terms of ‘grand bargains’ and the more ‘broad evolution of regional integration’ (Moravcsik and Schimmelfennig, 2009, p. 68). Whereas LI has predominantly been focussed on the delegation of competences, institutionalist authors have addressed the question what happens when supranational actors do not act according to the preferences of the MSs (Pollack, 2009, p. 132). It would therefore be unable to address more specific, day-to-day decision making practices in the EU. These questions have led to the hypothesis that agency autonomy varies over different policy fields (Pollack, 2009, p. 132). Partially addressing these criticisms, LI emphasises that there can be unforeseen changes in the domestic situation, thereby putting the MSs in an uneasy position

(Moravcsik and Schimmelfennig, 2009, p. 75). Moreover, supranational organisations can increase the autonomy of MSs: [T]hereby uncomfortably constraining governments in unexpected ways

(Moravcsik and Schimmelfennig, 2009, p. 75). Hereby, LI reaffirms its focus on agency.

Mechanisms

Moravcsik (Moravcsik and Schimmelfennig, 2009, p. 69) theorizes regional integration in a three-staged analysis: the formation of state preferences, the outcomes of inter-state bargaining, and the formation or adjustment of institutions. Regional integration can be understood as a series of rational choices made by governments that act under the different constraints and opportunities each level contains (Moravcsik, 1998, p. 8). These conditions have been specified in order to be able to generate expectations with regards to European integration. What expectations then, does the LI analysis generate about regional integration, and how does this framework respond to an attempt to generate expectations about disintegration?

(24)

The first part of the LI framework is the assumption that preferences are generated at the domestic level. The preferences of MSs are endogenous to the state; they are generated through a domestic political bargaining process where societal actors fight over specific policy goals, rather than that they depend directly on international circumstances or the integration process itself (Schimmelfennig, 2015, p.179). Societal actors are expected to determine preferences since governments seek re-election and are thus susceptible to pressure. Pressure is exerted directly through the mobilization of opposition and indirectly through punishment by voting (Moravcsik, 1993, p. 493).

Expectations about integration then, can be derived because LI assumes that the domestic power constellation tends to produce stable preference functions, specific to each ‘substantive issue’ (Moravcsik and Schimmelfennig, 2009, p. 70). Similarly Moravcsik and Schimmelfennig (2009, p. 78) argue that: ‘’The initial task in any LI analysis is to explain state preferences by understanding the structure of issue-specific domestic societal interests […]’’. This structure can be understood as the power constellation between domestic societal actors. For each issue, expectations can be derived for the: ‘’characteristic distributions of costs and benefits for societal groups, from which follow variations in patterns of domestic political mobilization, opportunities for governments to circumvent domestic opposition, and motivations for international

cooperation’’ (Moravcsik, 1993, p. 488). The groups which are important for preference formation then, struggle to influence the government. Which societal actors are then expected to be the most influential? LI expects that the strength of societal actors in securing that their interests are represented depends on the intensity of these interests. This is grounded in the ‘collective action problem’ that it is more rational to let others ensure that a collective good is reached in a large group. In this way it is easier to be a free-rider compared to a small group. Because of this problem, Olson (1965) states, diffuse majority interests will be overshadowed by concentrated minor interests since these concentrated actors are better able to overcome the ‘collective action problem’. Thus, is it more easy for an economic actor whose interests and resources are

(25)

concentrated, to make a strong stance in domestic preference formation. This leads to the expectation that the more organized societal actors are, and the more certain and unambiguous possible benefits are, the more likely economic preferences will dominate, vis-à-vis ideological preferences for example (Schimmelfennig, 2015, p. 179). Moravcsik (1993, p. 493) argued that producers are the societal actors that generally would be mobilized in policies that involved the direct regulation of goods and production processes; where the benefits would be unambiguous. Economic actors usually have very specific demands and hence are likely to better represented than other societal actors. The pressure on governments then results from producers: ‘’whose expected gains and losses reflect their competitive position on international markets, levels of intra industry trade, and the certainty of policy outcomes´´(Moravcsik, 1993, p. 495). With regards to the provision of socio-economic goods, their mobilization would be lower since the expected net benefits are less clear (Moravcsik, 1993, p. 491-494).

How can the process of preference formation be interpreted as to generate expectations with regards to disintegration? As with integration, disintegration is considered to be the result of domestic preference formation. As the mechanism of preference formation is grounded in the rationality assumption, disintegration can occur if this is in the interest of the dominant domestic societal actors. Similarly, pressure on governments stems directly from political mobilization or indirectly through voting. The expectation then, is that the domestic debate that results in domestic preferences determines disintegration. Preferences that are less ambiguous and which are highly organized have a comparative advantage to less intense preferences. Disintegration is then likely to happen if the outcomes of national debates favour disintegration.

(26)

The second part of the three-staged explanation of integration is the focus on the bargaining for substantive outcomes. It is unlikely that state preferences resulting from the first stage align. Outcomes of these negotiations are expected to be ‘pareto-efficient’; all states gain from an agreement (Schimmelfennig, 2015, p. 184). This is based on the assumption that decisions on treaties are taken on basis of unanimity and that no state will support an agreement if it does not reap benefit from it. In the case of OLP, the voting rule in the Council in case of Commission consent is by QM (if the Commission opposes a proposal under QMV, the Council votes by unanimity (Europa-nu.nl, 2017). Therefore the outcomes are not that easy to predict. Whereas integrative dynamics on treaties are directed by the governments that have a veto power, for other legislative decisions this is not the case. Therefore, the assumptions about how integration proceeds are different.

Whereas for treaties, integration often hinged on the Franco-German accord that could credibly threaten Great-Britain with exclusion (Webber, 2012), for other legislative proposals the relevant governments or coalitions of governments are more diverse. The dominance of France, Germany, and also Great-Britain however, has not entirely faded with the increase of the EU to 28 MSs. Their governments still represent the largest shares of votes and inhabitants, and especially when these governments form ‘opposing poles’ around which others may rally, they can be decisive (Webber, 2012). Roughly, outcomes can thus be interpreted as a reflection of veto-power constellations.

However, the bargaining process encompasses more since the possible outcomes of

bargaining processes are rarely clear and change during the process. The substantial outcomes of bargains then, are expected to reflect the constellation of states’ bargaining power

(Schimmelfennig, 2015, p. 184). Bargaining power results from asymmetrical interdependence; the uneven distribution of benefits from a potential outcome as opposed to alternative options (Moravcsik and Schimmelfennig, 2009, p.71). In general, the expectation is: ‘’… that states who are less vulnerable to interdependence gain less from integration’’ (Schimmelfennig, 20015, p.

(27)

184). Bargaining itself consists of ‘credible threats to veto proposals, to withhold financial side-payments, and to form alliances excluding recalcitrant governments’’ (Moravcsik, 1998, p. 3). For the European Union, more specifically, the expectation is that states that benefit the most economically, tend to compromise more in bargains, and states who benefit from the status quo have the important power of non-cooperation, giving them a position to compel others to concede more to their wishes (Moravcsik and Schimmelfennig, 2009, p. 71).

As is the case with integration, in order for disintegrative proposals to be adopted, a veto-coalition of MSs should be in favour of disintegration. It is expected that the most important MSs can function as ‘poles’ around which others can rally. The substantive bargaining process

determines the more specific outcomes of the eventual bargaining result. As with integration, these outcomes reflect the constellation of states’ bargaining power resulting from the uneven distribution of benefits from a potential outcome as opposed to alternative options. This part of the theory does not formulate specific criteria as to when to expect that a bargaining process results in disintegration, safe for the expectation that governments with a low vulnerability to interdependence benefit less from integrative schemes and hence are less prone to protect these. Attempts to change former agreements however, may meet resistance: ‘’attempts to ‘evade or question’ previous integration might trigger institutional response and undermine the basis of previous or other decisions (Moravcsik, 1995, p. 621).’’. Therefore, we also need to look at the choice of institutions to fully understand why (dis)integration happens.

 Cost benefit calculations

The third and last part of Moravcsik’s theoretical framework is the institutional choice. Not only do states bargain for the substantive terms of integration, they also negotiate the institutional design if they consider an institution to be necessary. The expectation that states sometimes agree to pool discretionary competences, is based on the idea that states must cope with ‘unintended, unforeseen, and often unwanted consequences’ (Moravcsik and Schimmelfennig,

(28)

2009, p. 72). The possible costs and benefits of integration are presumably weighed vis-à-vis the relevant alternatives. The benefits stem from the reduction of transaction costs of future

bargaining and the reduction of uncertainty about each other’s future preferences and behaviour by providing necessary information (Talberg, 2002, p. 26). The issue-specific variation of the institutional options preferred by governments then, is expected to reflect their concerns about other states’ future compliance (Moravcsik and Schimmelfennig, 2009, p. 72). In other words: governments try to shift the responsibility for the monitoring and sanctioning of compliance. Moreover, Vis-à-vis their domestic constituencies, these governments choose for the pooling of competences if they seek to circumvent domestic pressures to defect from an agreement (Moravcsik and Schimmelfennig, 2009, p. 72). It may be easier for them to avert responsibility if they are able to convincingly use a strategy of ‘tied hands’ to tackle their domestic pressure groups. States thus make a rational cost-benefit analysis of the benefits of integration vis-à-vis the alternative options to multilateral agreement. More generally, the relative success of MSs’ institutional choice reflects the institutional preferences of the MSs with the superior bargaining power (Schimmelfennig, 2015, p. 188). To summarize: it is expected that states opt for the pooling of discretionary competences when the substantive gains from integration are considered to outweigh the transaction costs of alternative options. The institutional choice then, is expected to reflect the institutional preferences of the most powerful MSs’ governments. In reverse, it should be expected that disintegration happens when the governments calculation of the gains from disintegration outweigh the costs of integration. Whereas for integration, the motivation for governments to create institutions is expected to stem from the reduction of the costs of future monitoring of compliance, for disintegration the expectation is that states opt for disintegration if the multilateral option of integration does not outweigh the costs of a disintegrated scheme. This can be a result of unexpected or changing costs of the integrative scheme relative to other options.

(29)

2.3 Comparative synopsis

This section’s purpose is threefold. First, it will include a discussion on the theoretical possibility of disintegration. Secondly, the discussed theories’ similarities and differences will be discussed. Third, some criticisms will be included in this discussion in order to contextualise the proposed explanation of disintegration.

Both NF and LI have understood EI as a project that is likely to widen and deepen and very unlikely to reverse (Rosamond, 2016, p. 866; Stone Sweet and Sandholtz, 1997, p. 310). This has led Webber (2012, p. 347) to classify these theories as unequivocal optimistic. Although authors of EI have discussed the possibility disintegration in the 1970’s already (Lindberg and Scheingold, 1970, p. 137; Schmitter, 1971), these comments have been marginal and unaccompanied by theoretical assumptions: ‘’Traditional theories of European integration are not particularly helpful with regard to determination of the mechanisms and conditions leading to disintegration (Cianciara, 2015, p. 56).

Rosamond (2016, p. 865) pointed at both scholars’ motivation for explaining EI, as well as a theoretical bias of LI and NF towards regarding the EU as institutionally resilient as possible

explanations for this absence. For NF the irreversibility of integration lies in the functional and economic interdependence between states. Although LI’s success is linked to the explaining of the stagnation of EI (Hoffman, 1966, Moravcsik, 1991), LI also regards integration as irreversible. As is the case with NF, LI supposes that this is the consequence of economic factors. LI supposes that, since bargaining takes place in a setting that is geared towards producing absolute gains (Moravcsik as paraphrased by Rosamond, 2016, p. 866), this ensures MSs’ future commitment to integration. As Moravcsik (1998, p. 493) phrases it then: ‘’…the irreversibility of integration [ lies ] in the evolution and adaptation of national preferences and institutions…’’. The end-result of integration for NF is a lasting set of supra-national institutions. Haas (1958, p. 16) expects that: ‘’The end result of a process of political integration is a new political community, superimposed over the pre-existing ones.’’. How these institutions would be allocated precisely, and how they would relate to each other, NF is unclear about (Schmitter and Lefkofridi, 2016, p. 7). For LI on the contrary, the end-result of

(30)

integration is no more than a ‘regime’ wherein states remain the decisive actors.

More recently however, attempts at a theoretical account of ED emerged especially on the possible morphology of such a process (Vollaard, 2008; Webber, 2012, Cianciara, 2015; Schmitter and Lefkofridi, 2016, Rosamond, 2016). Webber (2012, p. 359) argues that the European project is rather contingent upon domestic political backlashes stimulated by populism and the supporting role Germany plays in EI. Large crises such as economic and/or financial crises could therefore stimulate anti-EU sentiment and trigger large disintegrative stimuli. Cianciara (2015, p. 55) on the other hand acknowledges that disintegration can be a linear process whit differentiation as a consequence. Vollaard (2008, p. 21-22) points at the MSs cost of losing ‘voice’ as being too high, reaffirming LI expectations that institutions exist for the good of the MSs: Member States’ governments have still kept a tight grip on the most influential European institutions – they nominate most of the European Central Bank’s board, the European Commission, expert agencies, and steer Europolicies through the European Council, the Council of the European Union, and its expert committees. The Members States have become so much involved in Euro-level politics, that an escape or exit would be too expensive’’. Not all authors interpret explanations of ED as explanations of EI in reverse. On the contrary, argues Vollaard (2008, p. 8) bearing in mind Haas’ (1986) definition of integration: ‘’It immediately appears that European disintegration is not the same as the process of European integration reversed, because political actors do not necessarily direct their loyalties, expectations and political activities back to the national states but could shift these instead to (trans-national) regional authorities.’’. Although it may not necessarily be the case that political actors shift their focus back to the national states completely – indeed, it may be expected that once they have established transnational networks, the combined resources and shared goals guarantee the benefits of continued cooperation – , when discretionary competences are reallocated to the MS, political actors may be wise to continue to exert their influence on a lower level.

Although the end-result of disintegration might be different, its driving mechanisms may resemble the logics of integration in the sense that it may be rational for actors to pursue

(31)

disintegration. Based on the assumption of rationality, the possibility for disintegration would lie in it to outweigh the costs of alternative options. The implication of disintegration is thus present in theory, it has however never been made properly explicit: ‘’Countervailing forces, which constitute the other side of the equation, were of course always implicit in the neo-functionalist writings, but they were never subjected to the same degree of scrutiny and their importance was

underemphasised’’ (Tranholm-Mikkelsen 1991, 16). With regards to the underlying assumptions and mechanisms through which ED proceeds, these accounts draw from LI and NF nonetheless. As is the case for integration according to NF, some authors focus on the logics of disintegration: ‘’The ‘logic of disintegration’ consists of adherence to the symbols of sovereignty, the diversity of the MSs and other aspects of interdependence and security’’ (Tranholm-Mikkelsen 1991, 18). Vollaard (2008, p. 7) focusses more on MSs motivation: ‘’… the only option in functionalism for disintegration (or failed integration) is the state’’ (Vollaard, 2008, p. 7). The LI approach on the other hand suggests that the preferences of the most powerful MSs should converge for integration to occur, which has led several authors (Cianciara, 2015; Webber, 2012) to argue that disintegration is the consequence of diverging preferences of these states. The existing literature on disintegration does indeed draw from LI and NF to account for the possibility of ED. However, its expectations are not grounded in the core

assumptions of both theories, blurring the distinction between NF and LI and thereby their value as rigor hypotheses generating theories. For the sake of parsimony then, the theories of NF and LI are explored and reinterpreted as to account for disintegration.

The most important similarity in the two theories for this thesis, is that they both can be used to explain a reallocation of discretionary competences. They treat the EU as an external environment vis-à-vis which the relevant actors are positioned. The EU provides for different opportunities such as economic cooperation from which each party benefits. An important distinction needs to be made here between the assumption that actors act in a rational manner, and that the outcome of

integration is per definition rational. This thesis assumes the opposite, namely that it can be rational for actors to disintegrate the EU as well.

(32)

Rationality is a micro-assumption about behaviour which is used to make inferences about the meso-level of the EU. LI’s beginning was strongly motivated from an ambition to ground integration studies in a stronger theoretical foundation including micro foundational assumptions (Moravcsik and Schimmelfennig, 2009, p. 67). Theory, according to Moravcsik (1993, p. 477), requires micro-foundational assumptions about the nature of actors. This is necessary to be able to generalize empirical findings, or at least to make credible the expectations which are derived from research on integration. Moravcsik (1993) in particular stresses the failure of NF to ground its claims in such assumptions about why actors act the way they act. This is why NF is unable to make credible generalizations about EI: ‘’In this regard, neo-functionalism is both oddly apolitical and lacking in any aspiration to generality, in that it advances long-term predictions about the future of the EC without underlying, more specific theories that identify the decisive determinants of politicians’ choices among competing alternatives’’ (Moravcsik, 1993, p. 477). Indeed, NF is not known for its theoretical explanation of preference formation. In this sense, NF perhaps benefits more than LI from

institutionalist or governance work that differentiates between the differences between policy fields. However, both theories are susceptible to the same criticism: that mechanisms from LI and NF do not occur separately. Hence, an approach that encompasses a more nuanced understanding of EU politics, including formal procedures, would be more fitting to problems of EI, most notably governance theory and policy networks theory (Peters and Pierre, 2009; Peterson, 2009). What is useful about these criticisms is that, in EU politics, specific procedures and policy networks may be of importance. However, for this thesis, these aspects may function merely as resources for the

important actors in NF and LI which employ these in order to enhance their own position. Ontologically speaking, the focus on agency has generated some criticism amongst social constructivists who state that this focus cannot properly address socio-economic currents which may be conducive to perpetuating economic systems. As Cafruny and Ryner (2009, p. 232) argue: ‘’[T]he emphasis on formal decision-making in these accounts does not adequately address the underlying socioeconomic content of EMU or conflicts and contradictions posed for European society’’. This

(33)

choice of focus moreover, is argued to be obstructing a more ‘uneasy’ understanding of integration as proliferating an unequal economic system: ‘’[T]he concept of integration in the aforementioned sense is not entirely innocent. It implies that ‘’[E]conomic integration (the realm of economics) and social and political integration (the realm of international relations) are a priori idealized as the ‘rational’ and ‘general’, which can be distinguished from the ‘irrational’, ‘special interests’, and the realm of power politics that were associated with the old European interstate and inter-imperialist system’’ (Cox and Sinclair, 1996, p. 60-84 as paraphrased by Cafruny and Ryner, 2009, p. 222-223). This thesis is limited to this extent: it does not analyse the socio-economic relations in society. It rather focusses on agency, which has as the advantage that it can help allocate accountability. Both NF and LI are thus focused on the interplay between agency and structure. (dis)Integration is in this sense a function of the relevant actors’ interests, which are sometimes constraint or ‘upgraded’ at the EU level. Whereas for NF, the dividing line between MSs and the EU is blurred, and in the LI approach it is emphasised.

(34)

Chapter 3: Methods and hypotheses

The aim of this thesis is to explain why disintegration happens. In order to understand this, one specific event of disintegration is discussed: the reallocation of discretionary competences regarding the prohibition of specific GMCs for cultivation. In the previous chapter, theoretical presuppositions with regards to disintegration have been discussed. The purpose of this chapter then, is to provide the methodology with which the theoretical understanding of disintegration is linked to the empirical material that has been collected on this specific event. This methodology consist of a typology of the selected case and a discussion on its appropriate method of inquiry. In order to structure this line of inquiry, six hypotheses are introduced and operationalized. Furthermore, the choice and type of empirical material that is collected is discussed. Then, conclusively, a discussion on the advantages and disadvantages of the proposed method follows. A clear and disciplined presentation of the methodology is an important part of this thesis since, for a disciplined interpretive case study: ‘’The more explicit and systematic the use of theoretical concepts, the more powerful the application.’’ (Odell, 2001, p. 163).

3.1 Case selection

In order to generate expectations about when we can expect disintegration to occur, the case of the reallocation of GMC competences has been chosen to study. George and Bennet (2005, p. 17-18) define a case study as ‘’the detailed examination of a historical episode to develop or test historical explanations that may be generalizable to other events’’ wherein ‘a case’ is understood as ‘’an instance or a class of events’’. The event of a reallocation of discretionary competences was considered to be a common object of debate for EU leaders and societal actors for the first time during the French presidency of the Council around 2008 (Council, 2009) and was discussed formally in the EU during the years 2010 until its ratification in 2015. It’s implementation still continues today as MSs and societal actors respond to its implications. It forms a breach to the internal market in the sense that it differentiates the competitiveness of different MSs vis-à-vis other MSs. The merits of

(35)

different sorts of case studies analysing such crucial events, range from a defence and attempt of using it to test hypotheses and their reliability (Gerring, 2007), to an attempt to exploit its ability to enhance the understanding of a specific event and contribute to its theoretical understanding and its validity (Odell, 2001, p. 170). Whereas any good research attempts to combine both, it is important to understand that case studies always suffer from a trade-off between reliability and validity. Therefore, a case study design should address these limitations and discuss how these are mitigated through a

disciplined approach. A disciplined

interpretive case study (Odell, 2001, p. 163) applies a well-studied theory to a new object of study in order to explain an event. As LI and NF have been tested extensively, the relevance of this thesis would not increase with yet another test of their reliability. Instead it aims at generating a high validity of the described case. Moreover, it seeks to generate new theoretical expectations with regards to disintegration and thereby enhances the theoretical scope of LI and NF. Since a single case study has a smaller focus compared to multiple-case studies or large-N studies, it: ‘’…can represent a significant contribution to knowledge and theory building by confirming, challenging, or extending the theory’’ (Yin, 2013, p. 51). In contrast with a case study design aimed at testing hypotheses for their reliability, the disciplined interpretative case study aims at applying tested theory at a new phenomenon and hence shows that: ‘’…one or more known theories can be extended to account for a new event.’’(Odell, 2001, p. 163).

3.2 Method of inquiry

The method of inquiry in studying this crucial case is process tracing. Process tracing, the observance of consequential causal links in a process, is well suited for this thesis for a couple of reasons. It suits the theoretical framework since both theories assume causal processes. Whereas NF assumes an incremental process with small, sometimes unintended steps, LI considers disintegration to proceed via large decisions being made by governments. Both NF and LI use rational choice assumptions to make inferences to the international level, which gives process-tracing a comparative advantage over other methods on inquiry (Levy, 2008, p. 11). A condition for this is that the theory should provide

(36)

relatively precise predictions and has a low measurement error (Eckstein, 1975, p. 113-123).

Moreover, many qualitative methodologists consider one of the main tasks in case study research to be to generate as many testable implications of an hypothesis as possible (Levy, 2008, p. 3). The logic behind this method is implicitly based on the Bayesian assumption that the weight of the evidence is weighed relative to the prior set theoretical expectations (Levy, 2008, p. 12). Moreover, with regards to causality a disciplined interpretative case study can make us understand which expectations are important relatively to each other: ‘’…qualitative methods provide lesser precision in their

descriptions, claims about magnitudes of causal effects, and claims about the relative importance of different causes, than statistical methods’’ (Odell, 2001, p. 172). Therefore, the hypotheses’

expectations are to be assigned an expected valued influence for the hypothesis. Although assigning weight to different causes may be hard (Odell, 2001, p. 172), it gives more room for intra-theory comparison and provides for a more disciplined assessment of the theories.

3.3 Hypotheses and operationalisation

All hypotheses discussed below generate expectations with regards to the same independent variable. The dependent variable is disintegration. The concept of disintegration refers to the reallocation of discretionary competences. The different independent variables are introduced and operationalized in the hypotheses section below. As both theories assume that integrative dynamics are the result of societal debates, these shall form the basis of the empirical material.

 H.NF.1: Disintegration is likely to happen if practicalities speak in favor of diverging national approaches.

Practicalities that speak in favor of disintegration are understood as reasons for disintegration

stemming from the perceived interdependency between certain policy fields across countries. Practicalities that dictate disintegration, or functional spillback-linkages, can either stem from a perceived presence of interdependency (positive spillback-linkages), or a perceived relative absence thereof (negative spillback-linkages). In order to confirm this hypothesis two expectations are raised.

(37)

The first expectation is that there should be statements from decision-makers and/or societal actors, which refer to the high interdependency between integrative schemes in the EU. Moreover, these comments should address the harmful effects the object of disintegration may have for another initial integrative scheme. The rivaling expectation is that there should be statements from decision-makers and/or societal actors, which refer to the low interdependency between integrative schemes in the EU. These comments should address that the current pooling of discretionary competences obstructs uni- or multilateral relations with other sectors outside the integrative scheme. In order to disconfirm this hypothesis, practicalities should speak in favor of pooling discretionary competences. In the empirical material, there should be statements from decision-makers and/or societal actors, which refer to the high interdependency between policy fields across countries. Moreover, these actors should express that solutions for problems arising from an initial integrative scheme can be solved through further integration.

 H.NF.2: Disintegration is likely to happen if societal actors expect that a reallocation of competences better serves their interests.

Societal actors are originally understood to be domestic interest groups that exerted pressure on MSs

to promote their interest. As integration proceeds, they form international umbrella groups in which they align themselves with groups in other MSs with similar interests. Societal actors can here be understood as the transnational networks of societal actors who exert pressure domestically as well as internationally in order to promote their interests. NF expects that these societal actors have an influence in the (re)allocation of discretionary

competences. To confirm this hypothesis it is expected that empirical results show that internationally organized societal actors express dissatisfaction with the current allocation of

competences. Moreover, these statements should be accompanied with explicit statements by these actors in favor of a reallocation of GMC authorisation competences. In order for this hypothesis to be

(38)

disconfirmed, there should be statements from internationally organized societal actors in favor of the current allocation, or in favor of increased integration.

 H.NF.3: Disintegration is likely to happen if supranational actors do not wish to assume responsibility for a certain issue.

The supranational actor that has been regarded by NF as important in explaining EI, was initially the

high authority within the ECSC, and later the Commission (Niemann and Schmitter, 2009, p. 50). As the Parliament was included more and more in the decision-making process, the EP was ascribed similar influence on the integrative process. Assuming responsibility is related to the expected reputational gains or losses supranational actors take into account when making rational decisions. Reputation is a value judgement on the functioning of the institution as such, rather than merely on their policies. In order to confirm this hypothesis, empirical results should show statements by leaders in the Commission and/or the Parliament in favor a reallocation of discretionary

competences. The second expectation that is necessary in order for this hypothesis to be confirmed, is that the supranational organizations’ reputations should be harmed because of the integrative scheme. Strong empirical evidence then consists of explicit claims made by these organizations that current regulation is harmful to their reputation. A weaker form of confirmation would be empirical evidence that criticism on EU institutions encompasses more than just criticism on executed policy but focusses on the functioning or relevance of these institutions themselves.

 H.LI.1: Disintegration is likely to happen if the outcome of national debates speaks in favor of disintegration.

The outcome of national debates is synonymous to the preferences of MSs, except in that it emphasizes the process from which preferences originate. In order to confirm this hypothesis, empirical material should consist of statements made by governments in favor of disintegration. Furthermore, the formation of these preferences should confirm LI expectations regarding the high level of organization of domestic societal actors and low level of ambiguity of their goals. In order to

Understanding European disintegration