ARENA Working Papers
WP 01/18


An Organisational Approach to European Integration:

Outline of a Complementary Perspective

Morten Egeberg

Department of Political Science and ARENA

University of Oslo

Paper to be presented at the European Group of Public Administration

Annual Conference, University of Vaasa, 5-8 September 2001.

Research group: �Public Administration in an Integrated Europe�




An organisational approach to European integration focuses on individual actors� organisational context in order to account for their behaviour, interests and identities. Intergovernmentalists usually preclude any profound impact of EU institutions and organisations. Institutionalists (other than rational choice institutionalists), on the other hand, claim that EU institutions are able to shape and reshape individual actors� preferences and sense of belonging. Seen from an organisational perspective, however, institutionalists often fail to specify (and theorise) the organisational components that EU institutions contain. By focusing on organisational factors, it becomes evident that parts of the EU�s institutional set-up, e.g. its huge system of advisory committees, can not be expected to affect participants strongly. These structures simply impose too few demands on decision makers. However, other institutions, like the COREPER, even if it is highly present in the lives of policy makers, is organised in a way that makes it more likely that diplomats� basic ways of thinking and acting are sustained rather than seriously challenged.


The need for a complementary perspective

The purpose of this paper is to present an organisational approach to European integration. Although such a perspective may sometimes challenge ideas and insights advocated by other approaches, I find it more fruitful to consider it mainly as a complementary way to increase our understanding of what it is going on in the European Union, since I do not believe that any single theory is able to account adequately for everything. For example, liberal intergovernmentalism may be better equipped to explain �major turning points� in European integration (Moravcsik 1998:1) than how the system works on a daily basis (Peterson 1995; Krasner 1999:210). Different kinds of institutionalism ( cf. Aspinwall and Schneider 2000) may, on the other hand, do better in this respect. An organisational approach could increase our understanding of politics, policies and identity formation in settings that are organisationally well developed, but would have difficulties in explaining outcomes ofprocesses that take place in relatively �organisation-free� arenas. Even in highly structured contexts it would have difficulties in accounting for more specific outcomes, since that would presuppose detailed knowledge of actor intentions.����


Whereas Moravscik (1998) himself explicitly focuses on what he believes to be formative events in the history of European integration, the intergovernmental argument has been applied more generally on European integration for a long time (cf. Puchala 1999). From this perspective, policy-making at the European level is, in general, dominated by national governments whose interests and preferences are shaped and reshaped at the national level. Institutions, like the Commission and the Court, are managing co-operation among states by reducing transaction costs. Member countries are treated as unitary actors, thus, conflicts and cleavages at the European level are organised along (national) territorial lines. This intergovernmental view has been challenged by neo-functionalists (Haas 1958), various institutionalists (Bulmer 1994; Pierson 1996; Olsen 2000) and constructivists (Checkel 1999). According to these critics, institutions at the European level might play a much more significant role in the policy process, and they may be able to furnish participants with interests, preferences and identities, and even recast those already acquired at the national level. �Deliberationists� and constructivists would in particular highlight the role that arguing and persuasion might play in this respect (Eriksen and Fossum 2000; Risse 2000; Checkel, forthcoming). The extent to which member states actually comply with Community decisions was also put on the institutionalist research agenda. From an institutional perspective, implementation of EU legislation is seen as highly contingent upon national administrative traditions (Knill 1998; Sverdrup 2000).


From an organisational perspective, the extent to which institutions might impinge profoundly on people�s pre-established mind-sets and loyalties has to depend on how these institutions are organised. Settings that impose only relatively modest demands on decision-makers� attention, such as, for example, some Commission or Council groups, can not be expected to have the same impact as institutions to which individuals devote most of their present and future energy, like for example national ministries or Commission directorates. Some institutions may, however, involve policy-makers extensively without challenging their fundamental images and interests. It could, for example, be argued that COREPER is organised in a way that sustains and underpins rather than basically challenges the typical diplomat�s conception of the international society as one organised along territorial lines of cleavage. Diplomats in Commission settings would, on the other hand, become exposed to ways of thinking and expectations that are less compatible with their established belief systems, due to organisational characteristics of this particular setting. EU institutions like the Commission and the Parliament, both contain salient organisational components that may transform intergovernmental conflicts into sectoral, economic or social conflicts, thus eventually redistributing power in the system. Accordingly, an organisational focus may also provide a yardstick for assessing system transformation more accurately than what is offered by other approaches (Egeberg 2001).


�Organisationists� would also take a closer look at how governments hammer out and co-ordinate their national positions. Rather than taking for granted that member states are unitary actors, co-ordination should be seen as partly dependent upon how the process is organised at the national level (Kassim et al. 2000). Arguably, well co-ordinated positions that have emerged only after involving several institutions and procedures could be expected to be more �robust� at the European level than positions on which it is up to only one or two organisational units to decide. Furthermore, from an organisational point of view, it is far from obvious that an �authorised� national position is hammered out prior to European decision-making at all. This may be due to shortage of attention among those at the top who are supposed to endorse the policy position, or to unresolvable conflicts within national governments.


I proceed from here by presenting the approach�s key variables: organisational structure, organisational demography, organisational locus, and institutionalisation. The approach basically treats these variables as independent variables. However, we may also be interested in explaining and understanding organisational (re)structuring, recruitment, (re)locating and institutionalisation processes themselves. In that case, they are also treated as dependent variables. As will be seen, an organisational perspective focuses more on general features of organisations than on highly concrete legal procedures, like co-operation, co-decision, comitology, or QMV, as often focused upon in legal and rational choice institutionalism (cf. Aspinwall and Schneider 2000). Furthermore, an organisational approach offers an account of individual preference formation and change. Political analysis can not rely extensively on models that do not accommodate this vital aspect of political life. I then try to show how the organisational argument could be applied in the EU context. My illustrations are mainly drawn from the Commission and Council settings since these are the institutions with which I am myself most familiar.

Organisational key variables

Organisational structure. An organisational structure is a normative structure composed of rules and roles specifying, more or less clearly, who are expected to do what, and how (Scott 1981). Thus, the structure defines the interests and goals that are to be pursued, and the concerns that should be emphasised. The �relevance criteria� embedded in role expectations guide search processes, and bias information exposure. Thus, normative structures forge information networks for the development of agendas, alternatives and learning. Since a decision-maker is unable to attend to everything at the same time, and to consider all possible alternatives and their consequences (cf. �bounded rationality�), it seems to be a perfect match between her/his need for simplification on the one hand and the selection and filter that organisation provides on the other (Simon1965). The structure can therefore never be neutral, it always represents a mobilisation of bias in preparation for action (Schattschneider 1975:30).


What reasons then do we have to expect that people will comply with organisational norms from the moment they enter an organisation? First, they may feel a moral obligation to do so. Modern cultures, emphasising impersonal relationships and �rationalised� codes of conduct in organisational life, assist individuals at separating their private interests from those that should be catered for in their capacity as employees or representatives. Second, they may find compliance to be in accordance with their self-interest. Organisations are incentive systems that inform members at lower levels of their potential career prospects, thus inducing them to adopt autonomously to role expectations and codes of conduct. And managers may apply rewards and punishments in order to achieve obedience. Finally, and third, social control and �peer review� by colleagues are supposed to make deviant behaviour less likely. Thus, these mechanisms do not imply that organisational members give up their private interests from the moment they enter an organisation. However, personal preferences are put aside and are thus supposed to be of minor importance in explaining organisational behaviour. Even if the mechanisms fail, it could be argued that participants would be unable to define and operationalise their genuine private interests in any meaningful and coherent way in most issue areas. One obvious exception to this could, however, be decision processes that might impact more directly on their career prospects, for example, reorganisation processes (Egeberg 1995).


I now turn to various dimensions of organisational structure. The size, the sheer number of roles that are to be filled, may indicate its capacity to initiate policies, develop alternatives, or to implement final decisions. Horizontal specialisation expresses how different issues and policy areas, for example transport and environmental protection,are supposed to be linked together or de-coupled from each other. Those areas that are encompassed by the same organisational unit are more likely to be co-ordinated than those that belong to different units (Gulick 1937). However, in a hierarchy (i.e. a horizontally and vertically specialised organisation), separation of issues at lower levels only means that co-ordination responsibility is moved up to higher echelons, thus making it more likely that a process becomes politicised (Egeberg 1999a). According to Gulick (1937) there are four fundamental ways in which tasks may be distributed horizontally among units, namely in relation to territory, purpose (sector), function, or clientele served. If, for example, an organisation is internally specialised according to the geographical area served, it is expected to induce spatial perspectives and encourage policy-makers to pay attention primarily to particular territorial concerns and need for �intra-local� policy coherence. In this case, the structure reflects the territorial composition of the system and focuses attention along territorial lines of cleavage. Organisations based on a purpose principle, on the other hand, are supposed to foster sectoral horizons among decision-makers and policy standardisation across territorial units.


The structure may express whether co-ordination is supposed to be hierarchical, as within national ministries and Commission directorates, or �collegial�, as in many national cabinets and the College of Commissioners. The structure may be more or less centralised or decentralised, as found in unitary versus federal state systems. It may be more ambiguous or loosely coupled than other structures, thus facilitating innovative behaviour and extensive policy dynamics (March and Olsen 1976; Richardson 1996; Heritier 1999; Hood 1999). Enduring tensions and unresolvable conflicts may also be dealt with more intelligently through ambiguous designs (Olsen 1997).


Overlapping structures designate a situation in which a person who occupies a particular role in one organisation is expected to fill a particular role also in another organisation. For example, ministers are affiliated to particular ministries and the cabinet as such at the same time, and parliamentarians have to play party roles as well as committee member roles. Multiple roles of this kind may entail serious cross pressure on the incumbents. In overlapping structures one of the structures may be more dominant than the other(s). A primary structure is a much more �demanding� structure than a secondary structure. Affiliation to a primary structure means that a person is expected to use most of her or his time in a particular organisation. The organisation is her/his main employer. Secondary structures, on the other hand, usually engage people only on a part-time basis. The typical setting is a committee system. Modern systems of governance co-ordinate policies extensively across levels and sectors in committees. Thus, participants become exposed to new agendas, alternatives, actors, obligations and rewards. We therefore expect that committees, like other organisational arrangements, might affect the perspectives, interests and identities of those who attend. However, the impact will be less profound than in organisations to which persons have a primary affiliation. If overlapping structures contain the same or compatible principles of specialisation, conceptions and mind-sets will be underpinned and sustained when decision-makers move from one role to another. Are, however, these principles different and incompatible, we could expect perspectives and beliefs to be challenged when participants move from on setting to another.


Organisational demography. According to Pfeffer (1982:277) demography refers to the composition, in terms of basic attributes such as age, sex, ethnicity, education and length of service of the social entity under study. Such factors are supposed to impact on decision behaviour, although the strength of potential effects have to depend on characteristics of the organisational structure, for example how �demanding� it is (Meier and Nigro 1976; L�greid and Olsen 1984). Even more, a wide variety of socialisation experiences are not relevant to policy disputes and thus are unlikely to reveal a representational linkage (Selden 1997:65). One may say that the demographic perspective emphasises the effects that flows of personnel (where people come from, their present and future careers) might have on their decision behaviour. Whereas the effects of organisational structure are thought to occur without any socialisation of personnel, the impacts of demographic factors are closely related to socialisation. Socialisation usually means that values, norms and role expectations have become internalised in individuals. New recruits arrive �pre-packed� with images and attitudes acquired over the years in particular social, geographical and educational settings. With increasing length of service in a particular organisation, they may, however, become resocialised. Socialised organisational members identify themselves strongly with a particular organisation, and are supposed to advocate its interests �automatically� in the sense that these interests are taken for granted and legitimate without further consideration. Arguably, the extent to which an organisation has to rely on external control mechanisms (incentives and sanctions) depends on the extent to which decision-makers have become socialised within that same organisation.


Considered as individual attributes, only length of service can, in a strict sense, qualify as a real organisational factor among the demographic variables mentioned. However, this becomes different if we instead deal with proportions of a given organisational �population� that come from, for example, different regions or professions. Clusters, or �enclaves�, seem to make it more likely that particular group interests might be pursued (Selden 1997).��


Organisational locus. The physical dimension of organisational life has not been emphasised very much in the literature (Pfeffer 1982:260-71). However, most organisations are located in particular places and buildings. First, features of location and physical space segregate personal lives and their associated role conceptions and identities from organisational roles and identities. Second, overlapping organisational structures that are separated in space (and then often time) provide cues for evoking different roles and identities, while concentration in space (and then often time) makes it more probable that role perceptions and identities are carried over from one unit into another (March 1994:70-73). Third, physical distance within and between government buildings seems to affect contact patters and co-ordination behaviour (Egeberg 1994). In short, organisational locus, like organisational structure, creates boundaries that focus decision maker�s attention and assist them in coping with a complex reality. Processes involving considerable uncertainty, unpredictability and surprise require information exchange via face-to-face contacts and group conversation. Thus, such processes are in a sense highly locus dependent (J�nsson et al. 2000:186).


Institutionalisation. From an organisational point of view all institutions are organisations, not all organisations are, however, institutions. Institutionalisation is a dimension of organisations that adds important characteristics. Thus, the present tendency to classify all kinds of rules, regimes and organisations as institutional phenomena has given us a poorer concept of institution. According to Selznick (1957), institutionalisation necessarily takes time. It means that organisations are growing increasingly complex by adding informal norms and practices. This increased complexity stems from the organisation�s continuous interaction with its environments (e.g. information exchange, recruitment), and provides it with a broader repertoire for handling major challenges. These informal norms and role expectations are impersonal in the sense that they exist independently of the concrete individuals who happen to be in the organisation at different points in time. Thus, the informal structure should not be mixed with norms and role expectations associated with the organisational demography.


To become a real institution, however, Selznick (1957:17-22) argued that the �grown-up� and complex organisation also had to be infused with value beyond the technical requirements of the task at hand. By this he meant that an organisation acquires a self, a distinctive identity, involving the taking on ofvalues, ways of acting and believing that are deemed important for their own sake. For the individuals who participate directly in it, an organisation may acquire much institutional value, yet in the eyes of the larger community the organisation may be readily expendable. Thus, arguably, from a political perspective, organisations become real institutions as they come to symbolise the community�s aspirations, its sense of identity. Real institutions embody societal values, and strive to impose those same values on society. Institutionalisation could mean that not only particular organisational structures and informal norms become infused with value and meaning, but also a particular demographic composition of the organisation, for example as regards professional groups, and also the place and building associated with the organisation (Goodsell 1988). Thus, it is probably no coincidence that revolting groups often occupy the presidential palace or the parliamentary building. Such action may be interpreted primarily as symbolic action; it doesn�t aim at making political decisions in the first place.


A broad context of�understood� meanings may represent a considerable aid to communication in institutions. It may create energy that increases performance and co-ordination, and be of special importance in times of crisis or threat (Selznick 1957:18; Brunsson and Olsen 1993:5). Thus, compared to an organisation, an institution probably does not have to rely on external control mechanisms (rewards and punishments) to the same extent. An other implication of institutionalisation deals with the possibilities for deliberate reform and reorganisation. �An organization that does take on this symbolic meaning has some claim on the community to avoid liquidation or transformation on purely technical or economic grounds� (Selznick 1957:19). The inherent robustness of institutions now seems widely acknowledged in the literature. Changes which accord with the institutional identity are supposed to be carried out as a matter of routine. However, sudden big changes which violate this identity are rare, and when they do occur they are assumed to be the result of serious performance crises (Brunsson and Olsen 1993:5-6; March and Olsen 1989). An alternative interpretation is that identity-challenging reforms are indeed implemented, but only in a highly history-, path-dependent and distorted way (Christensen og L�greid 2001).


Quasi-institutionalisation. There may exist an alternative to the historical path to institutionalisation. All organisations and institutions find themselves within �institutionalised environments�, that is environments composed of legitimate models of what is seen as good and modern organisation, procedures and recruitment practices (Meyer and Rowan 1977). So, maybe then, young organisations could institutionalise faster by adopting already historically developed forms that are seen as highly appropriate for equivalent organisations. There certainly are models �out there� of how executives, legislatures, courts and central banks could be properly structured, staffed and housed.

Applying the argument on the European Union - some examples  

National co-ordination of EU-related policies. Metcalfe (1994) found that the extent to which member states co-ordinate their EU-related policies vary considerably. Later, it has been well documented that this variation in actual co-ordination behaviour is partly due to how the process is organised. In, for example, France and the UK the co-ordination process is hierarchically structured and concentrated at the Prime Minister�s Office, while in, for example, Austria and Germany, where ministerial autonomy is a key principle of government, ministers or their representatives can act with considerable independence in the European arena (Kassim 2000). A decentralised, federal structure may add to this picture of a more loosely coupled system (Derlien 2000). However, although a clear distinction emerges between member states, care should be taken not to overestimate the extent to which �the best co-ordinated� act coherently on EU-related matters. In France as well as in the UK other ministries, like the foreign and finance ministries, may be allowed to hammer out their positions with considerable autonomy in certain policy fields (Kassim 2000:249). Co-ordination behaviour also seems to vary within countries dependent upon the structural position of those supposed to be co-ordinated. Those in government agencies vertically separated from ministries, who often attend Commission expert committees, are less involved in co-ordination than their colleagues in the minsitries who often attend Council working parties (Schaefer et al. 2000; Trondal 2001a).In general, from an organisational point of view, overlapping structures and the ecology of simultaneous decision processes may create serious shortage of attention, in particular among those who are expected to guide and instruct others on which policy positions they should take in external arenas (March and Olsen 1976). It is not uncommon, therefore, that national representatives on EU committees themselves have to formulate their instructions (Lewis 1998; Trondal 2001a). Since these officials may be heavily involved in EU level committees (Beyers and Dierickx 1998; Schaefer et al. 2000), it is reason to believe that, when fulfilling this task, they do not remain unaffected by their EU level roles (Lewis 1998; Kassim 2000:239).And, finally, it is not necessarily so that there are national interests to be pursued in all issue areas. Sometimes they might eventually be discovered during processes at the EU level.


What difference then does it make for EU-level decision-making if policies are well or badly co-ordinated at the national level? Arguably, national positions that have emerged only after involving several institutions and procedures could be expected to be more �robust� at the European level than positions on which it is up to only one or two organisational units to decide. Those who are badly co-ordinated bring more latitude to the bargaining (or arguing) table and will probably find it more convenient to calibrate or even recast their original positions. However, meagre co-ordination at the national level may have more far-reaching implications too: it may facilitate the development of new lines of cleavage at the European level where the dominant pattern of conflict along national borders becomes increasingly challenged by sectoral, economic and social divides across territorial boundaries.


The Commission and its personnel. The role that national interests might play in Commission decision-making is a highly contentious and enduring issue. Most authors assign some weight to the �national connection�, in particular at the levels of the cabinets and college (Coombes 1970; Cini 1996; Nugent 2001). It has, however, also been stated that the Commission is even permeated by national interests, and acts as an important forum for competition between them (Peterson 1999:59). What can be said from an organisational perspective in this respect? Starting with the Commission services, they are obviously composed of nationals who are supposed to be �pre-packed� with national experiences, norms and values. One could argue that the informal national quota system regulating recruitment in relation to the population size of the member states (Spence 1994) might serve to legitimate the evocation of national identities and policy paradigms within the EU administration. Empirical studies show that Commission officials serve as points of access for their compatriots (Michelmann 1978; Egeberg 1996). They also seem to bring with them different administrative styles, associated with, for example, a southern versus a northern European culture (McDonald 1997). Nationality also impacts on their beliefs and attitudes on intergovernmentalism versus supranationalism (Hooghe 1999a), on a Weberian versus a consociational Commission (Hooghe 1999b), and on socialism versus capitalism in Europe (Hooghe 2000). Although these personal attitudes may be seen as some sort of paradigms, belief systems or conceptual lenses that might somehow make a difference in a given decision situation, they are, nevertheless, of a relatively general nature. To become relevant in a given decision context, they have to be operationalised, and they have to pass several potential organisational filters. First, considering organisational demography, officials may become resocialised. For example, it has been shown that the longer an official stays in the Commission, the more he or she is likely to become supranationalist (Hooghe 1999a). Routinely, steps are also taken in order to avoid national clusters or enclaves to develop. Thus, staff immediately below or above a given senior post should be of a different nationality (Spence 1994), and the divisions (�units�) are multi-nationally composed (Egeberg 1996). Second, for most officials the Commission is their primary organisational affiliation. Most posts are permanent (Spence 1994), and they are mainly grouped according to purpose (sector) or function, thus making it less likely that the incumbents will focus on territorial (national) concerns as such (Egeberg 1996). Studies do seem to reveal that the structural (DG) attachment of officials probably is the best predictor of their decision behaviour (Cram 1994; Egeberg 1996; McDonald 1997; M�rth 2000). Also, interviews and a survey conducted among national officials who participate in EU-level policy-making show that Commission officials are seen as acting mainly independently from particular national interests (Egeberg 1999b, Schaefer et al. 2000; Trondal 2001b).


Concerning the College of Commissioners there are aspects of its organisational structure that might be conducive to enhancing the importance of demographic background factors like nationality. In particular, the fact that commissioners are nominated by member governments for a limited number of years may have this effect. In addition, their private offices (cabinets) have traditionally been mainly staffed by their compatriots. On the other hand, role expectations are unambiguous; instructions from outside �the house� should not be taken. Reforms that have assigned to the President of the Commission and the European Parliament a more important role in composing the college may further contribute to �autonomising� the Commission from member state control. Multi-national staffing of cabinets probably work in the same direction. Finally, since the commissioners are embedded in overlapping structures, namely the college and the directorates general, the relocation of commissioners over to their respective services may have strengthened their sectoral identities at the expense of their collegial and national affiliations (Egeberg 2001).(1)However, only careful empirical examination can reveal whose interests commissioners actually pursue. This topic is highly underresearched indeed.


Has the Commission become institutionalised so that the robustness, legitimacy and sense of mission often associated with institutions are to be reckoned with? One interpretation is that efforts at quasi-institutionalisation, thus by-passing the historical path of institutionalisation, have indeed been made. Administrative jargon and components, like, for example, cabinets and concours (competitive examinations), have obviously been borrowed from French administration, probably the most prestigious bureaucracy found among the early member states. British and Danish accession may have accelerated the need for legitimating elements that are seen as integral to their administrative culture. For example, too much cabinet involvement in the appointment of top officials are, from their point of view, inappropriate practices. When the Committee of Independent Experts also unveiled instances of nepotism in their report (that in fact sacked the Santer Commission), it seems quite understandable that Vice President Kinnock, in his reform programme, declared appointments could be made solely on the basis of merit and experience.(2)


With a history that goes back to the establishment of the High Authority of the European Coal and Steel Community in 1952, the Commission has existed long enough to have become an institution �in the real way�. Obviously, it is today something more than a formal organisation. It contains informal norms that are important for the operation of the organisation, for example the national quota system (Spence 1994). Although several cultural traits can be traced back to the administrative cultures of the member states, authors increasingly seem to point out that a distinguishable Eurocrat-culture is slowly emerging (Abeles et al. 1993; McDonald 1997). The Commission � �the House� as it is referred to colloquially and affectionately by its staff � has developed its own ethos and a strong esprit de corps (Shore 2000:127). An important part of its mission, namely to act independently from particular national interests, is also highly acknowledged by national officials who deal with Commission officials in expert committees and Council working parties (Egeberg 1999b; Schaefer et al. 2000; Trondal 2001b). However, since the Commission more clearly than any of the other EU institutions symbolises the EU�s departure from the traditional intergovernmental organisation, its legitimacy in the wider political space may be more problematic. While EU-sceptics would like to see the Commission more as a second secretariat for the Council, pro-integrationists prefer to perceive it as the European government. The Commission�s somewhat unclarified role is also reflected in some of the scholarly literature which portrays the EU executive as dual, where the Council and the Commission share the responsibilities of government (Hix 1999:21). However, although the proper role of the Commission is still contested, its existence is not on the agenda. Thus, it has become an institution in the sense that it is taken for granted.��


Governance by committees. At the EU level we find both primary and secondary organisations. The Commission, for example, is the primary affiliation of the Commissioners as well as of the officials in the services. Thus, we would expect their interests and identities to be shaped mainly by the Commission setting, and even more so for the officials who hold permanent posts than for the Commissioners. The committees of different kinds under the Commission and the Council represent secondary arenas of overlapping structures for national and Commission officials. Accordingly, preferences and basic role perceptions are supposed to change only partly, although additional concerns and loyalties might emerge. The fact that EU meetings are separated in time and space from the daily activities of most national officials makes the evocation of new perceptions more likely. Particular places, buildings and symbols, like the blue flag with the golden stars, may gradually become associated with a particular code of conduct. However, since some committees are more active than others the extent to which change actually takes place may vary across groups. COREPER, for example, meets weekly and is composed of national officials who also live in Brussels for several years. The location facilitates extensive informal interaction across nationalities, and could be thought to make the COREPER setting particularly conducive to preference change. A sense of extraordinary collective responsibility and supranational loyalty seem indeed to have complemented national allegiances in this case (Lewis 1998; Trondal 2001a). The working parties at the level below are also comprised of several people from the Permanent Representations in Brussels, although they are supplemented considerably by officials brought in from the national capitals. However, as expected, also among genuine �part-timers� in Council and Commission committees one can observe behavioural and attitudinal traits that may be interpreted as having a supranational flavour (Kerremans 1996; Joerges and Neyer 1997; Beyers and Dierickx 1998; Egeberg 1999b; Trondal and Veggeland 2000; Trondal 2001a; 2001b).


The organising principles underlying EU committees are more or less compatible with the principles embedded in the national institutions from which the committee participants originate. The Council�s basic principle of specialisation is geography (territory) in the sense that each participant (except the Commission representative) represents a particular national government. The sectorally specialised ministerial meetings and working parties modify to a certain extent the intergovernmental logic of the institution, though. The Commission, on the other hand, divides its work primarily according to sector or function. At all levels, including the preparatory expert committees, participants are not expected to represent their country of origin. The Council structure, and particularly COREPER, can be said to be compatible with the organising principle expressed in the existence of a Foreign Ministry and its Permanent Representations whose mission is to represent a particular country. Thus, although frequent interaction among diplomats in the Council may lead to enhanced collective responsibility and reciprocity, one could argue that their territorial role perceptions and identities acquired in their primary institutions are sustained rather than profoundly challenged by the organisational characteristics of the Council. In the Commission setting, on the other hand, diplomats would face an incompatible environment that is supposed to challenge their established perspectives, for example by focusing attention along sectoral lines of cleavage rather than territorial ones. Those from national sector ministries are supposed to make the opposite experiences. Their beliefs will be further underpinned by the Commission structure while put under a certain pressure in the Council. For example, Jacobsson (1999) observed that the Swedish accession to the EU entailed an increased demand for Swedish policy positions. Thus, through EU participation sector experts became more aware of their national identities. Thus, we assume the extent to which preferences and identities may be moved or reshaped depends on the organisational characteristics of the institutional setting.��

What organisations tells us about system transformation

There are many proposals out there on how the transformation of Europe�s political order might be properly portrayed. The emerging system has, for example, been described as multi-level, interwoven, deliberative, made up of concentric circles, and supranational. No doubt these labels grasp parts of reality. Many of them are, however, relatively vague and do not necessarily point to aspects that are peculiar to the European Union. There are, for example, important elements of supranationality and �multi-levelness� in other international organisations as well. My interpretation is that what is taking place is no less than a transformation of the state system in Europe as it has been known since the second half of the seventeenth century. This order was organised along territorial lines so that only interstate conflicts were brought to the fore at the European level. The institutional building blocks were the territorial states and intergovernmental conferences and organisations. These organisations may have facilitated collective problem-solving among national governments, however, one could argue that such organisations have sustained rather than transcended the existing state system and its inherent pattern of conflict and co-operation. The reason may be found in the way these organisations have been specialised; their structure reflects the territorial composition of the system and thus underpins the system�s basic line of cleavage. In the EU the Council most clearly fills this role, although its spatial logic is challenged by sectorally and functionally specialised ministerial meetings and working parties. The fundamental new thing in post WWII Europe is the appearance of organisations and institutions at the European level that, due to their specialisation, focus conflict along non-territorial lines, for example, sectoral, functional, social and economic lines.The European Commission, Parliament, Court of Justice and Central Bank all allow (and mostly oblige) participants to depart from national role conceptions. These organisations, arguably, frame politics differently, emphasising policy makers sense of belonging to directorates, interest groups, political parties or professions rather than to particular nation states. Roles, identities and patterns of conflict that are well known at the national level are thus transmittedand expanded to EU arenas. It is a well established insight from political science that expanding (�socialising�) conflict, or refocusing lines of cleavage, entails new allocations of power so that new sets of winners and losers are produced (Schattschneider 1975). This development may be fuelled by national compartmentalisation and the tendency to let prime ministers play an increasingly central role in EU matters at the expense of foreign ministries (Kassim 2000:236). While pursuing national interests are at the core of foreign ministries� mission, prime ministers are, arguably, more used to launch party political programmes, for example, on the role of the public sector in the economy.��


Patterns of conflict in political life are thus here seen as tightly linked to institutions and organisations. There are many conflicts and potential conflicts out there. Some of these are organised into politics, some are organised out. Institutions discriminate among conflicts, they channel conflict and do not treat all conflicts impartially. For example, it is not given from nature that the main line of cleavage at the European level since the second half of the seventeenth century has been found among states. It is indeed a result of the organisation of political power in Europe. Centralised state powers may have, more or less, tolerated differentiated patterns of conflict and loyalty at the national level, but have at the same time claimed subordination of sub-national conflicts and allegiances on the international scene. Nevertheless, there are instances where political parties, interest organisations and sub-national regions have tried to ally with their counterparts across national borders in order to reallocate power (cf., for example, socialist trade unions and parties). At the European level, however, conflict and co-operation among nations have been clearly dominant.


Based on an organisational understanding we are now able to formulate more precise criteria according to which the degree of system transformation might be assessed: The extent to which sub-territories are politically integrated into a larger system is reflected in the extent to which the interests of these sub-territories are expressed organisationally at the centre. Thus, in a highly integrated political system, non-territorial principles of organisational specialisation have taken clear precedence over the territorial principle at the centre (Egeberg 2001). As in unitary states, the institutional configuration at the centre does only marginally reflect the territorial composition of the system. In weakly integrated systems, on the other hand, like traditional intergovernmental organisations, the overarching governance structure is geographically specialised at the top level. Similar to the federal state, but unlike the unitary state and the intergovernmental organisation, the EU embodies a certain balance between different specialisation principles. In contrast to the federal state, though, in the EU non-territorial components do not seem to have take any precedence over territorial ones so far. However, my interpretation is that, over time, reform efforts and actual changes have gradually strengthened non-territorial principles of organisational specialisation at the EU level (Egeberg 2001).


An organisational approach questions some of the basic assumptions made by intergovernmentalists. The unitary actor hypothesis is challenged by emphasising the highly specialised, horizontally as well as vertically, character of modern national governance systems. The existence of hierarchical devices for co-ordination varies across countries, but also across different arenas within countries. Executive politicians and senior officials, who are supposed to hammer out national policy positions, are parts of overlapping structures that tend to cause serious shortage of attention. Badly co-ordinated preferences (if preferences are formulated at all) may be more conducive to profound alteration at the EU level than those that have emerged only after cumbersome clearance processes involving a huge number of actors.


Like neo-functionalists, institutionalists (other than rational choice institutionalists) and constructivists, �organisationists� consider preference formation and change endogenous to their models. However, organisational analysts would find it necessary to specify the organisational setting in a much more precise way in order to clarify the conditions under which preference alteration might take place, and in what direction changes may occur. This should, however, be done without having to rely heavily on rather concrete, legal categories, as is often the case in rational choice institutionalism. Instead, the focus should be on dimensions of a more abstract and genericnature that may be theoretically linked to particular role perceptions, identities, patterns of conflict, and decision processes. And, finally, an organisational approach provides a yardstick for measuring system transformation in a relatively consistent way across systems of governance.



1 See also European Voice 8-14 February 2001, p. 12.

2 Press statement by Vice President Neil Kinnock, 29 September 1999.���������



Abeles, M., Bellier, I. and McDonald, M. (1993) Approche Antropologique de la Commission Europeenne`. Executive Summary. Brussels: The European Commission.

Aspinwall, M. and Schneider, G. (2000) �Same menu, separate tables: The institutionalist turn in political science and the study of European integration�, European Journal of Political Research 38: 1-36.


Beyers, J. and Dierickx, G. (1998) �The working groups of the Council of the

European Union: supranational or intergovernmental negotiations?�, Journal of

Common Market Studies 36: 289-317.


Brunsson, N. and Olsen, J.P. (1993) The Reforming Organization. London: Routledge.


Bulmer, S. J. (1994) 'The governance of the European Union: a new institutional

approach', Journal of Public Policy 13: 351-380.


Checkel, J. (1999) �Social construction and integration�, Journal of European Public Policy 6: 545-60.


Checkel, J. (forthcoming) �Why comply? Social learning and European identity change�, International Organization 55:


Christensen, T. and L�greid, P. (2001) �A transformative perspective on administrative reforms�, in T. Christensen and P. L�greid (eds), New Public Management. The Transformation of Ideas and Practice. Aldershot: Ashgate.


Cini, M. (1996) The European Commission. Leadership, Organisation and Culture in the EU Administration. Manchester: Manchester University Press.


Coombes, D. (1970) Politics and Bureaucracy in the European Community, London: George Allen and Unwin.


Cram, L. (1994) �The European Commission as a multi-organization: social policy and IT policy in the EU�, Journal of European Public Policy 1:195-217.


Derlien, H.-U. (2000) �Co-ordinating German EU policy. Failing successfully?�, in H. Kassim, G. B. Peters and V. Wright (eds), National Co-ordination of EU Policy-making. Oxford: Oxford University Press.


Egeberg, M. (1994) �Bridging the gap between theory and practice: the case of administrative policy�, Governance 7: 83-98.


Egeberg, M. (1995) �Bureaucrats as public policy-makers and their self-interests�, Journal of Theoretical Politics: 7: 157-67.


Egeberg, M. (1996) 'Organization and nationality in the European Commission

services', Public Administration 74: 721-735.


Egeberg, M. (1999a) �The impact of bureaucratic structure on policy-making�, Public Administration: 77: 155-70.


Egeberg, M. (1999b) �Transcending intergovernmentalism? Identity and role perceptions of national officials in EU decision-making�, Journal of European Public Policy 6: 456-74.


Egeberg, M. (2001) �How federal? The organisational dimension of integration in the EU (and elsewhere)�, Journal of European Public Policy 8 (Forthcoming)


Eriksen, E.O. and Fossum, J.E. (eds) (2000) Democracy in the European Union. Integration through Deliberation. London: Routledge.


Goodsell, C.T. (1988) The Social Meaning of Civic Space. Studying Political Authority through Architecture. Lawrence: University Press of Kansas.


Gulick, L. (1937) �Notes on the theory of organization. With special reference to government�, in L. Gulick and L. Urwich (eds), Papers on the Science of Administration, New York: Institute of Public Administration, Columbia University.


Haas, E.B (1958) The Uniting of Europe. Stanford: Stanford University Press.


Heritier, A. (1999) Policy-Making and Diversity in Europe. Escape from Deadlock. Cambridge: Cambridge University Press.


Hix, S. (1999) The Political System of the European Union. Houndmills: Macmillan Press.


Hood, C. (1999) �The garbage can model of organization: Describing a condition or a prescriptive design principle?�, in M. Egeberg and P. L�greid (eds), Organizing Political Institutions. Essays for Johan P. Olsen, Oslo: Scandinavian University Press.


Hooghe, L. (1999a) �Supranational activists or intergovernmental agents? Explaining political orientations of senior Commission officials to European integration�, Comparative Political Studies 32: 435-63.


Hooghe, L. (1999b) �Consociationalists or Weberians? Top Commission officials on nationality�, Governance 12: 397-424.


Hooghe, L. (2000) �Euro-socialists or Euro-Marketeers? EU top officials on capitalism�, Journal of Politics:


Jacobsson, B. (1999) �Europeiseringen och statens omvandling�, in K. Goldmann, J. Hallenberg, B. Jacobsson, U. M�rth and A. Robertson, Politikens Internationalisering, Lund: Studentlitteratur.


Joerges, C. and Neyer, J. (1997) �Transforming strategic interaction into deliberative problem-solving: European comitology in the foodstuffs sector�, Journal of European Public Policy 4: 609-25.


J�nsson, C., T�gil, S. and T�rnqvist, G. (2000) Organizing European Space, London: Sage.


Kassim, H, (2000) �Conclusion: The national co-ordination of EU policy: Confronting the challenge�, in H. Kassim, B.G. Peters and V. Wright (eds), The National Co-ordination of EU Policy. The Domestic Level, Oxford: Oxford University Press.


Kassim, H, Peters, B.G. and Wright, V. (eds) (2000) The National Co-ordination of EU Policy. The Domestic Level. Oxford: Oxford University Press.


Kerremans, B. (1996) 'Do institutions make a difference? Non-institutionalism,

neo-institutionalism, and the logic of common decision-making in the European Union', Governance 9: 217-40.


Knill, C (1998) �European policies: The impact of national administrative traditions�, Journal of Public Policy 18: 1-28.


Krasner, S (1999) �Logics of consequences and appropriateness in the international system�, in M. Egeberg and P. L�greid (eds), Organising Political Institutions, Oslo: Scandinavian University Press.


Lewis, J. (1998) �Is the �hard bargaining� image of the Council misleading? The Committee of Permanent Representatives and the local elections directive�, Journal of Common Market Studies 36: 479-504.


L�greid, P. and Olsen, J.P. (1984) �Top civil servants in Norway: key players - on different teams?� in E.N. Suleiman (ed.), Bureaucrats and Policy-Making, New York: Holmes & Meier.


March, J.G. (1994) A Primer on Decision Making. How Decisions Happen, New York: The Free Press.


March, J.G. and Olsen, J.P. (1976) Ambiguity and Choice in Organizations, Bergen: Scandinavian University Press.


March, J. G. and Olsen, J.P. (1989) Rediscovering Institutions. The Organizational Basis of Politics, New York: The Free Press.


McDonald, M. (1997) �Identities in the European Commission�, in N. Nugent (ed), At the Heart of the Union. Studies of the European Commission, Houndmills: Macmillan Press.


Meier, K.J. and Nigro, L.G. (1976) �Representative bureaucracy and policy preferences: a study in the attitudes of federal executives�, Public Administration Review 36: 458-69.


Metcalfe, L. (1994) �International policy co-ordination and public management reform�, International Review of Administrative Sciences 60: 271-90.


Meyer, J.W. and Rowan, B. (1977) �Institutionalized organizations: Formal structure as myth and ceremony�, American Journal of Sociology 83: 340-63.


Michelmann, H.J. (1978) �Multinational staffing and organizational functioning in the Commission of the European Communities, International Organization 32: 477-96.


Moravcsik, A. (1998) The Choice for Europe. Social Purpose and State Power from Messina to Maastricht, London: UCL Press.


M�rth, U. (2000) �Competing frames in the European Commission � the case of the defence industry and equipment issue�, Journal of European Public Policy 7: 173-89.


Nugent, N. (2001) The European Commission, Houndmills: Palgrave.


Olsen, J.P. (1997) �Institutional design in democratic contexts�, Journal of Political Philosophy 5: 203-29.


Olsen, J.P. (2000) �Organising European institutions of governance. A prelude to an institutional account of political integration�, in H. Wallace (ed), Whose Europe? Interlocking Dimensions of Integration, London: Macmillan.


Peterson, J. (1995) �Decision-making in the European Union: Towards a framework for analysis�, Journal of European Public Policy 2: 63-93.


Peterson, J. (1999) �The Santer era: the European Commission in normative, historical and theoretical perspective�, Journal of European Public Policy 6: 46-65.


Pfeffer, J. (1982) Organizations and Organization Theory, Boston: Pitman.


Pierson, P. (1996) �The path to European integration. A historical institutionalist analysis�, Comparative Political Studies, 29: 123-63.


Puchala, D.J. (1999) �Institutionalism, intergovernmentalism and European integration: A review article�, Journal of Common Market Studies 37: 317-31.


Richardson, J. (1996) �Policy-making in the EU. Interests, ideas and garbage cans of primeval soup�, in J. Richardson (ed), European Union: Power and Policy-making, London: Routledge.


Risse, T. (2000) ��Let�s argue!� Communicative action in world politics�, International Organization 54:


Schaefer, G.F., Egeberg, M., Korez, S. and Trondal, J. (2000) �The experience of member state officials in EU committees: A report on initial findings of an empirical study�, Eipascope 2000/3: 29-35. Maastricht: European Institute of Public Administration.


Schattschneider, E.E. (1975) The Semisovereign People, Hinsdale: The Dryden Press.


Scott, W.R. (1981) Organizations. Rational, Natural, and Open Systems, Englewood Cliffs: Prentice-Hall.


Selden, S.C. (1997) The Promise of Representative Bureaucracy. Diversity and Responsiveness in a Government Agency, Armonk: M.E. Sharpe.


Selznick, P. (1957) Leadership in Administration. A Sociological Interpretation, Berkeley: University of California Press.


Shore, C. (2000) Building Europe. The Cultural Politics of European Integration, London: Routledge.


Simon, H. A. (1965) Administrative Behavior, New York: The Free Press.


Spence, D. (1994) �Staff and personnel policy in the Commission�, in G. Edwards and D. Spence (eds), The European Commission, Harlow: Longman.


Sverdrup, U. (2000) Ambiguity and Adaptation. Europeanization of Administrative Instituions as Loosely Coupled Processes. Ph.D. dissertation, Oslo:ARENA.


Trondal, J. (2001a) Administrative Integration across Levels of Governance. Integration through Participation in EU Committees. Oslo: Department of Political Science.


Trondal J. (2001b) �Is there any social constructivist-institutionalist divide? Unpacking social mechanisms affecting representational roles among EU decision-makers�, Journal of European Public Policy 8:1-23.


Trondal, J. and Veggeland, F. (2000) �Access, voice and loyalty. The representation of domestic civil servants in EU committees�, Oslo: ARENA.