ARENA Working Papers
WP 01/5

 

Beyond the EU Membership-Non-Membership Dichotomy?

Explaining Supranational Identities Among EU Decision-Makers

Jarle Trondal

ARENA

University of Oslo

Introduction[1]

Increasingly, studies of European integration acknowledge the vertical blurring of governance levels in Europe (e.g. Schaefer 2000; Wessels 1998). European integration constantly penetrates and transforms the core elements of the governing dynamics of European nation-states. Processes of European integration and national adaptation (Europeanisation) have increasingly become two flips of the same coin. However, until recently, the study of European integration largely ignored Europeanisation of national government systems. The current study aims at shedding light on how government levels become blurred due toprocesses of Europeanisation. The present contribution studies the extent to which national civil servants evoke supranational allegiance. Supranational allegiances resemble identifications with EU level institutions � e.g. towards particular EU committees and/or towards the EU more generally. Suggestively, if supranational allegiances exceed pre-established national and sectoral identifications, the individual actors become Europeanised. Ultimately, if several national civil servants �go native� in Brussels, the national central administrations get increasingly Europeanised.

How can we best account for supranational allegiances? [2] Several suggestions are depicted in the literature (Beyers 1998; Hooghe 1999; Rosamond 2000; Sandholtz and Stone Sweet 1998). Three hypotheses on supra-nationalism are highlighted here. First, supra-nationalism is triggered by the EU membership of each nation-state. Arguably, national government officials from EU member states evoke supranational allegiances more strongly than do officials from non-member states. The current study compares officials from the two EU member states Denmark and Sweden, and officials from the non-member-state (or more correctly, the EEA state) Norway. Second, several scholars assume that intensive and sustained participation on supranational institutions, like EU committees, accompanies supranational allegiances amongst the participants (Haas 1958; Lewis 2000; McDonald 1998; Niemann 1998; Trondal 2001). This assumption derives from the old neo-functional hypothesis on political spill-over processes and elite socialisation in supranational networks. In order to test this hypothesis, the current study compares national officials participating on EU committees with different degrees of intensity and duration. The data set applied covers national officials attending Commission expert committees and Council working parties. Finally, some authors have argued that national co-ordination may hamper the emergence of supranational allegiances amongst the national civil servants (Kassim, Peters and Wright 2000; Moravcsik 1998). Supra-nationalism is fostered by the general lack of national co-ordination; the lack of written mandatory mandates and instructions from the national political-administrative leadership. In the present study we compare national officials attending EU committees with varying degrees of ex ante instructions and mandates.

The empirical analysis builds on survey data on 160 Scandinavian government officials employed at the ministry and agency level with various experiences from EU committees. Additionally, 47 qualitative face-to-face interviews were conducted on a systematic sample of the above officials. The current article goes like this: The second section outlines an institutional approach on supra-nationalism and derives three different, however supplementary, empirical hypotheses from this approach: (i) Supra-nationalism reflects the EU membership of the respective nation-states; (ii) supra-nationalism is due to intensive and sustained participation on EU committees amongst national civil servants; and finally (iii) supra-nationalism reflects the lack of national co-ordination. The third section of the article presents the data in greater detail. The fourth section provides an empirical analysis, shedding light on the relative explanatory power of the three hypotheses on supra-nationalism suggested above.

Towards three hypotheses on supra-nationalism

Whereas past literature on European integration draw primarily on neo-functionalist and intergovernmentalist accounts, an �institutionalist turn� has been observed in the 1990s (Andersen 2000). However, this turn borrows heavily from past theories on European integration. Even more, this turn has moved the study of European integration more in the direction of analysing national adaptation to the European integration processes. Still, based on old neo-functionalist insights, recent institutionalist accounts on European integration emphasise that national officials involved in EU decision-making processes may shift orientations, attitudes, role perceptions and institutional identifications (e.g. Jensen 2000: 75; Lewis 1999; Tranholm-Mikkelsen 1991; Trondal 2001). An institutional perspective focuses on how supra-nationalism may reflect the institutional embeddedness of nation-states, of individual officials, and of particular decision processes. Assuming that institutions �matter�, three analytical takes are outlined below, providing three different and supplementary explanations of supranational allegiances amongst EU committee participants.

EU membership �matters

First, supranational allegiances evoked by national civil servants may reflect the EU membership of the nation-state. The enactment of supranational identifications amongst individual decision-makers is thus accounted for by the EU membership versus non-membership distinction. Some have suggested that processes of Europeanisation may be observed in EU member-states only. �[W]ith EC-membership [states] will start moving in the direction of Europeanisation and convergence whereas countries outside the EC � will not follow this direction until they have gained full membership� (Wessels and Rometsch 1996:357 � emphasis added). As such, this represents a macro-level institutional approach to supra-nationalism. EU member-states have institutionalised and rutinised linkages to the various EU institutions, and thus forced to comply with community regulations more extensively than non-member-states. EU member-states also become socialised into �community methods� to a larger extent than non-member-states. Hence, we would expect government officials from EU member-states to evoke supranational identifications more strongly than are officials from non-member-states.

However, two important caveats are warranted: First, the EU membership versus non-membership is a continuum rather than a dichotomy. Nation-states may have different forms of affiliation to the EU, as well as different degrees of interaction with different Union bodies (Egeberg and Trondal 1999; Stubb 1996). For example, while Denmark has made reservations concerning the Common Foreign and Security Policy, Norway have decided to provide troops to a European rapid reaction force. The distinction between insiders and outsiders of the Union becomes blurred and ambiguous. Consequently, the EU membership versus non-membership distinction becomes an ambiguous explanatory tool-kit because the extreme endpoints on that variable are seldom reached in real life situations. Second, individual civil servants are embedded in different institutional environments. The EU membership variable represents an aggregate institutional environment in political-administrative life. Arguably, more immediate institutional environments mould individual identifications more strongly than do more aggregate institutional contexts. Hence, when explaining supra-nationalism amongst non-aggregate entities like civil servants, it is probably wiser to rely on explanatory tools of a non-aggregate character. In the following, two non-aggregate explanations of supranational allegiances are suggested: First, the intensity and length to which national civil servants participate on EU committees, and second, national co-ordination processes occurring prior to meetings within EU committees.

Intensive and sustained participation on EU committees

Does supra-nationalism reflect intensive and sustained participation on EU committees amongst national civil servants? Moving from the macro-contextual explanation suggested above, the current approach to supra-nationalism focuses on institutional arguments at the more meso-level. Most national civil servants have several institutional affiliations simultaneously. Some of these affiliations are primary to them, some more secondary. Accompanying diverse institutional affiliations, �human behaviour is considered � multifaceted, exhibiting different logics� (Knill and Lenschow 2001: 5). Hence, a multifaceted and multi-layered conception of identity underpins this study. However, contrary to Haas (1958), I argue that supranational allegiances are supplementary to pre-established national and sectoral identifications, not a replacement.

Institutions provide cognitive frames, opportunity structures, normative assessments, physical boundaries for sense-making, action, institutional allegiances and role perceptions (March and Olsen 1989). �Institutions do not affect simply the strategic calculations of individuals, but also their most basic preferences and very identity� (Knill and Lenschow 2001: 8). Institutions remind actors daily about the appropriateness of different decisions, the set of cognitive frames possible to select, the Pareto-optimal choices available, and the goodness-of-fit between environmental standards and local adaptations (Peters 1999). Focusing on how institutions change the identities of individual actors, a new-institutional perspective is applied focusing on actors as bounded rational. Limits on the cognitive capacities of individuals render them vulnerable to and sensitive towards institutional values, norms, goals and demands (Simon 1997). Institutions provide stimuli to the actors, triggering particular responses in return.

Different institutions matter in different ways (Gulick 1937; Hammond 1990). However, institutions also matter to different degrees (Egeberg and Trondal 1999). Building on identity theory, actors with life-long careers and with few competing organisational affiliations are likely to evoke uni-dimensional and uni-layered identities. Actors who have participated loyally and faithfully in one singular organisation for several years, and who generally interact intensively with fellow organisational members are likely to identify strongly with that particular organisation. Yet, actors normally have several partially contending organisational affiliations simultaneously. Competing logics of appropriateness� and contending cognitive schemes strive constantly for attention. Hence, actors may evoke a multitude of different identities that fit only at the margins. Multiple institutional affiliations may trigger at least three different responses amongst the actors: (i) they might attend to one overarching identity that precedes all others, (ii) they might construct different allegiances that co-evolve over time, or (iii) they might construct different identities that are acted on sequentially in different institutional contexts. I argue that the latter scenario is most likely amongst national civil servants attending EU committees. Hence, coping with identity conflicts need not be synonymous with resolving those conflicts.

National central administrations and EU committees are intersecting circles of government.These institutions represent spheres of activity in which identities may be moulded and remoulded (Pescosolido and Rubin 2000: 55). In the following will be argued that EU committees may be conceived as secondary institutional affiliations to national government officials, supplementing pre-existing affiliations towards national ministries, agencies, professions, corporations, etc. In this study, national ministries and agencies represent primary institutional affiliations to the civil servants, whilst EU committees are considered secondary institutional affiliations. National civil servants attending EU committees are thus highly pre-packed and pre-socialised before attending the EU committees. Most national officials are only temporarily and ephemerally affiliated towards the EU committees. Henceforth, pre-established national allegiances and professional identifications are likely to take primacy over supranational allegiances. National administrations and EU institutions might not develop into more than imperfect integrated orders.

Still, under certain conditions supranational identifications are more likely to be evoked by EU committee participants. That is, if the participants generally attend the EU committees intensively and for protracted periods of time. Protracted and intensive exposure towards certain organisations render it likely that actors �go native�, increasingly identifying with the organisation attended (Haas 1958). However, in the current study identity is of organisational character only, and not of personal character. Institutional identities do not have the same depth, scope, durability and inertia as, for instance, sexual and parental identities. Institutional allegiances are more mouldable, changeable and possible to construct and reconstruct than for example parental identities. Hence, national officials participating extensively on EU committees are likely, over time, to evoke new supranational allegiances, supplementing pre-established national and sectoral allegiances. Hence, supra-nationalism stems from the sheer intensity and length to which national government officials attend EU committees.

National co-ordination mechanisms

The third hypothesis argues that supra-nationalism stem from the lack of genuine national co-ordination. Suggestively, lack of co-ordination prior to EU committee meetings may accompany the enactment of supranational allegiances amongst the EU committee participants. Those officials attending EU committees are not free riders, nor are they solely trustees. Officials attending Council working parties and COREPER meetings are shown in particular to be guided by well co-ordinated national positions (Schaefer, Egeberg, Korez and Trondal 2000; Trondal 2001). Participants to Council working parties are often equipped with written mandates of mandatory character (Trondal 2001). These instructions are often the result of inter-ministerial co-ordination processes, often involving the Ministry of Foreign Affairs (MFA) as a vital interlocutor and gardener of the �national interests�. The MFA might be regarded as the last stronghold of the nation-state. Hence, the extent to which the co-ordination role of the MFA is weakened in the national decision-making processes preceding EU committee meetings, the committee participants become more likely to evoke supranational loyalties. Lack of written mandates from national government institutions might thus propel the emergence of loyalties towards EU institutions. Not having clear mandates or instructions to follow, the room for additional supranational allegiances to develop increases. Not being reminded of their �national missions� on a daily basis in Brussels, the actors more easily loose sight of the nation-state as their primary locus of loyalty. Even more, actors participating on a daily basis within EU committees, frequently meeting and interacting with fellow committee participants, are more likely to evoke supranational allegiances, especially if these actors have few and weak national mandates.

Nevertheless, one caveat is warranted. National co-ordination is mostly geared towards governing the actual decision-making behaviour of the EU committee participants (Kassim, Peters and Wright 2000; Metcalfe 1994). For instance, Norwegian officials attending Commission expert committees are intended to draw up instructions considering the following aspects: background and status on the particular dossier, existing national regulations, Norwegian interests on the subject matter, the present bargaining situation, and finally administrative, economic, budgetary and juridical consequences (The Prime Minister�s Office 1997). Beyond this, national instructions seldom entrust the participants with particular loyalties, identities or allegiances. Hence, national co-ordination may not have a direct causal effect on supra-nationalism. The effect may be indirect, through the decision-making behaviour chosen by the actors.

Data and methodology

This study builds on two sources of empirical data: One survey and several qualitative interviews. First, from summer 1998 until spring 1999 a survey was conducted amongst Danish, Norwegian and Swedish government officials with experiences from EU committees. The sample selected was systematic, excluding officials without such experiences. No statistical generalisations are possible on the basis on our analysis. Still, theoretical generalisations are possible (Yin 1993). 203 questionnaires were distributed by post to these officials, whereby 160 answered. The response rate is thus 79 per cent. [3] Additionally, 47 qualitative face-to-face interviews were conducted amongst a systematic sub-sample of the above survey sample. Owing to the fact that most EU committees deal with dossiers of a highly technical nature, most of the officials selected to the survey study and to the interviews are employed at the agency level, and employed in medium rank positions. Hence, our sample of respondents largely mirrors the average EU committee participant (Institute of Europ�ische Politik 1987; Pedler and Schaefer 1996; Schaefer, Egeberg, Korez and Trondal 2000).

Systematic samples, like ours, often accompany certain biases. First, our study concerns officials employed in two different policy sectors: the environment sector and the occupational health and safety sector. These two policy areas were selected due to the fact that the environmental sector is heavily integrated in the EU aquis and the EU integration process, whereas the field of occupational health and safety is less so. Hence, selecting these two policy sectors increase the likelihood of selecting civil servants with widely different experiences from EU committees. Second, this study covers officials from three �small� European countries: Denmark, Norway and Sweden. One bias that may accompany this selection is that officials from small states tend to be more supranational than officials from larger states (Beyers 1998; Hooghe 1999). Finally, this study has been conducted in three unitary states. Civil servants from political systems where political authority is vested in national central institutions are shown to be less supranational than officials from federal states (Hooghe 1999). Hence, two biases, which partly are at odds with each other, are represented in our study.

Three layers of identification

Actors embedded in several institutions, at different levels of governance, might arguably enact several partially competing institutional allegiances. The respondents in our survey were asked to identify the institutional allegiances evoked most strongly when attending EU committees. Table 1 compares the relative strength of supranational allegiances versus pre-established identifications.

Table 1 Proportion of national civil servants evoking different institutional identities when attending EU committees (%).

 

 

Supranational allegiances:

Allegiance towards the EU committees attendeda��������.

 

35

Allegiance towards the EU as a whole�������������

25

�An �esprit de corps� emerges over time in the EU committees�b��..

41

 

Pre-established allegiances:

Allegiance towards my own department������������.

 

 

78

Allegiance towards my own ministry/agency as a whole������

92

Allegiance towards my own government������������

83

Allegiance towards my own profession/educational background��...

44

Allegiance towards my own policy sector�����������...

63

Mean N�������������������������

120

a) Variables 1, 2, 4, 5, 6, 7 and 7 combine values 1 and 2 on the following five-point scale: to a very great extent(1), to a fairly large extent (2), both/and (3), to a fairly small extent (4), and to a very small extent (5).

b) Variable 3 combines values 1 and 2 on the following five-point scale: agree totally (1), agree to a fairly great extent (2), agree somewhat (3), agree to a fairly small extent (4), and disagree totally (5).

Table 1 clearly reveals that national civil servants retain pre-established institutional allegiances when attending EU committees. Supranational allegiances are only supplementary to pre-established identifications. Those identities enacted most strongly are: allegiances towards their own ministry and agency (92 per cent), allegiance towards their own government (83 per cent), and allegiances towards their own department (78 per cent). Allegiances towards their own policy sector (63 per cent) and their educational background (44 per cent) are less prominent. Hence, institutional allegiances seem to reflect the immediate institutional environments embedding national civil servants (cf. below). Moreover, pre-existing institutional identifications are evoked more extensively than supranational allegiances (cf. Egeberg 1999; Schaefer, Egeberg, Korez and Trondal 2001; Trondal 2000; Trondal and Veggeland 2000). According to one Swedish agency official,

�I feel strongest allegiance to Sweden, but develop a certain amount of loyalty to the committee. Still, this loyalty never exceeds the loyalty to the [national agency]� (Author�s translation). In a similar vein, one Swedish agency official argued that, �it is hard to change ways of acting and thinking� (Author�s translation).

However, it seems that attendance on EU committees affects the scope, or width, of the supranational allegiances evoked: EU committee participants seem to identify more strongly with their own EU committee (35 per cent) than with the EU as a whole (25 per cent). Similarly, 41 per cent of the EU committee participants report that an �esprit de corps� emerges in the committees over time. According to one Norwegian agency officials attending Commission expert committees,

�I feel a certain sense of belonging to the group, or to individuals of the groups, especially to Nordic colleagues whom we meet occasionally in other contexts� (Author�s translation).

Henceforth, the EU committees may be considered more immediate institutional environments than the EU as a whole. However, whereas identifications with the EU committees attended and with the EU as a whole is strongly interrelated (Pearson�s r .54**), none of these variables correlate significantly with the tendency of feeling an �esprit de corps� within the EU committees. Hence, identification towards EU institutions (EU committees and the EU as a whole) seems to represent a different aspect of supra-nationalism than the �esprit de corps� variable. Whereas the two former variables represent personal and individual identifications towards particular EU institutions, the latter variable may partly represent personal experiences and partly observations of the general atmosphere and milieu in the committees. Hence, identifications towards particular EU institutions are perhaps a more adequate operationalisation of supra-nationalism than the emergence of an �esprit de corps�.

Next, does supra-nationalism merely reflect the EU membership of the nation-states, the intensity and length to which individual civil servants attend EU committees, or the general lack of ex ante national co-ordination mechanisms? These three hypotheses are dealt with separately below.

Does EU membership really �matter�?

The effects of EU membership have been revealed empirically in several recent scholarly contributions (Egeberg and Trondal 1999; Hanf and Soetendorp 1998; Jacobsson and Sundstr�m 2000; L�greid 2000; M�ny et al. 1996; Wessels and Rometsch 1996). Table 2 reveals the extent to which EU membership accompanies supranational allegiances amongst national civil servants attending EU committees.

Table 2 Proportion of national civil servants evoking supranational identities, distributed by country (%).

 

Allegiance towards the EU committees attendeda

Allegiance towards the EU as a whole

�An �esprit de corps� emerges over time in the EU committees�b

 

 

 

Mean N

Norwegian officials���.

Swedish officials����.

Danish officials�����

16

48

36

8

36

16

35

42

45

29

48

35

 

112

a)Variables 1 and 2 combine values 1 and 2 on the following five-point scale: to a very great extent (1), to a fairly large extent (2), both/and (3), to a fairly small extent (4), and to a very small extent (5).

b)Variable 3 combines values 1 and 2 on the following five-point scale: agree totally (1), agree to a fairly great extent (2), agree somewhat (3), agree to a fairly small extent (4), and disagree totally (5).

Table 2 provides fairly strong indications that EU membership matter as to the enactment of supranational allegiances amongst EU committee participants. Most significantly, fewer Norwegian officials seem to evoke allegiances towards the EU as a whole and towards the EU committees compared to Danish and Swedish officials. Similarly, Danish and Swedish civil servants feel more strongly that an �esprit de corps� emerges within the EU committees compared to the Norwegian civil servants. However, the differences between the Danish and Swedish officials on the one hand, and the Norwegian officials on the other is far more extensive related to the two former dependent variables. Even more, fewer Norwegian officials attend EU committees compared to Danish and Swedish officials (Trondal 2001). Hence, EU membership might accompany the enactment of supranational identifications amongst a greater number of Danish and Swedish national civil servants than Norwegian officials. As such, supranational allegiances amongst individual Danish and Swedish civil servants seem to reflect the EU membership of Denmark and Sweden, respectively.

However, the observed effects of EU membership may potentially contain spurious relationships. Hence, when analysing the evocation of supra-national allegiances at the individual level we might need to go beyond the aggregate effects of the EU membership versus non-membership distinction.

Beyond the EU membership � non-membership dichotomy

Intensive and sustained participation on EU committees.

As the EU becomes increasingly differentiated, allowing for various levels of �enhanced co-operation� and different forms of affiliation towards the Union, the membership versus non membership dichotomy becomes a less adequate analytical tool-kit. First, individual institutional affiliations towards EU committees may accompany supranational allegiances amongst individual civil servants. Arguably, officials attending EU committees fairly intensively and for protracted periods of time are more likely to enact supranational allegiances than officials devoting little time and energy participating on these committees. Table 3 reveals that this hypothesis merits attention.

Table 3 Does supranational identities reflect the intensity and length to which national officials attend EU committees? A multiple OLS regression analysis (beta)a.

 

Allegiances towards the EU committees attendedb

Allegiances towards the EU as a whole

�An �esprit de corps� emerges over time in the EU committees�c

Affected by the EU/EEA agreement�������

 

.02

 

.26**

 

-.03

Time devoted to EU related issues generally�����

 

-.06

 

.01

 

.06

Time devoted participating on Commission expert committees and Council working parties�����

 

 

 

.02

 

 

 

-.02

 

 

 

.23**

Number of EU committees attended�������...

 

-.09*

 

.05

 

.15**

Number of meetings in EU committees last year���

 

.14*

 

.04

 

.19**

Do you give oral presentations during EU committee meetings?��...

 

 

.08*

 

 

.13*

 

 

.11*

Face-to-face contacts outside formal EU committee meetings���.

 

 

.33**

 

 

.05

 

 

-.14*

Contacts by phone, e-mail, fax, etc. outside formal EU committee meetings���.

 

 

.05

 

 

.08

 

 

.12*

Number of years having attended EU committees�..

 

.07

 

.00

 

-.06

*) p .05��������� **) p .01���������������������������� R2 = .15��������������������������� R2 = .12��������������������������� R2 = .15

a)       Independent variables: The values of variable 1: To a very great extent (1), to a fairly great extent (2), somewhat (3), to a fairly little extent (4), to a very little extent (5). The values of variable 2: very great proportion (1), fairly great proportion (2), both/and (3), fairly little proportion (4), very little proportion (5). Variable 3 combines values 1, 2 and 3 on the following five-point scale: very great proportion (1), fairly great proportion (2), both/and (3), fairly little proportion (4), very little proportion (5). Variable 4 is dichotomised as follows: one committee (2), two committees or more (1). Variable 5 is dichotomised as follows: one meeting (2), two meetings or more (1). Variable 6 combines value 1 and 2 on the following five-point scale: very often (1), fairly often (2), sometimes (3), fairly seldom (4), very seldom/never (5). Variables 7 and 8 have the following values: very often (1), fairly often (2), sometimes (3), fairly seldom (4), very seldom/never (5). Finally, variable 9 is dichotomised as follows: between 1994 and 1998 (2), before 1994 (1).

b)       Dependent variables: Variable 1 and 2 applies the same value labels as table 2. Hence, the dependent variables are dichotomised. This poses no problems for the regression analyses due to the univariate frequencies on each original dependent variable is evenly distributed.

������ Dependent variable: cf. b.

Table 3 applies nine different empirical proxies measuring the sheer intensity and length to which national officials attend EU committees. These measures range from the less demanding task of being affected by the EU/EEA agreement at the one end of the spectrum towards participating on EU committees and having informal face-to-fact contacts with fellow committee participants on the other. Most notably, table 3 reveals that national officials participating fairly intensively and lengthy on EU committees seem to evoke supranational allegiances more strongly than officials participating with less intensity. Most notably, when studying the magnitude of the different beta coefficients revealed in table 3, supra-nationalism seems most strongly affected by informal face-to-face contacts amongst fellow committee participants (.33**), by being affected of the EU/EEA agreement (.26**), and finally, by the time devoted participating on EU committees (.23**). Even more, giving oral presentations during EU committee meetings seems to relate significantly positive on all three operational measures of supra-nationalism presented in table 3.

Only two deviant observations are detected in table 3: Officials attending a great deal of EU committees seem to identify with the EU committees less strongly than officials attending fewer EU committees (-.09*). This relationship, however, is hardly significant. The second deviant observation revels that officials having a great deal of informal face-to-face contacts with fellow committee participants outside formal committee meetings report that an �esprit de corps� emerges to a lesser extent compared to officials having fewer face-to-face contacts (-.14**). This beta coefficient is also fairly weak. Substantively, this negative relationship may reflect the fact that officials having many informal contacts outside the EU committees become less strongly exposed to the �esprit de corps� emerging within each committee. By and large, however, the above observations render it likely that intensive and sustained participation on EU committees accompanies supranational allegiances amongst the participants.

National co-ordination mechanisms.

The third hypothesis suggests that supranational allegiances reflect the general lack of national co-ordination and governmental steering. Suggestively, national officials that are not bounded and tied up by written instructions from the national political-administrative leadership, but rather act like trustees, are more likely to attend to supranational identities than officials who are controlled, instructed and governed by national mandates. Table 4 provides no clear evidence that may support this hypothesis. [4]

Table 4 Does supra-nationalism reflect the general lack of national co-ordination? A multiple OLS regression analysis (beta)a.

 

Allegiances towards the EU committees attendedb

Allegiances towards the EU as a whole

�An �esprit de corps� emerges over time in the EU committees�c

Doing clearances with other central administrative institutions before attending EU committees������

 

 

 

 

-.08

 

 

 

 

-.00

 

 

 

 

-.03

Doing clearances with the MFA��������..

 

-.15*

 

-.16*

 

-.02

Paying heed to signals from the political leadership nationally��.

 

 

-.03

 

 

.15*

 

 

.15*

�I co-ordinate with the MFA or with other central co-ordinating units���..

 

 

.21*

 

 

.05

 

 

.03

�My �position� has been co-ordinated with all relevant ministries���...

 

 

-.12*

 

 

.10

 

 

(.12*)

�I have clear instructions as to what �positions� to pursue in EU committees�

 

 

.05

 

 

-.14*

 

 

.01

*) p .05��������� **) p .01������������������������� R2 = .04���������������������������� R2 = .07���������������������������� R2 = .07

a)       Independent variables: Variable 1 combines value 1 and 2 on the following five-point scale: to a very great extent (1), to a fairly great extent (2), somewhat (3), to a fairly little extent (4), to a very little extent (5). Variable 2 has the following values: Yes (1), No (2). Variable 3 combines value 1 and 2 on the following five-point scale: very important (1), fairly important (2), both/and (3), fairly unimportant (4), very unimportant (5). Variable 4, 5 and 6 has value 1 on the following three-point scale: always or fairly often (1), half of the time (2), seldom or never (3).

b)       Dependent variables: Variable 1 and 2 applies the same value labels as table 2.

������ Dependent variable: cf. b.

First and foremost, table 4 unravels an equal number of predicted patterns and deviant observations. Supportive to our hypothesis, officials evoking supranational allegiances tend to do relatively few clearances with the MFA before attending EU committees (-.15* and -.16*), tend to co-ordinate poorly with all relevant ministries (-.12*), and have unclear instructions as to what �positions� to pursue during EU committee meetings (-.14**). These observations suggest that supra-nationalism reflects the general lack of ex ante national co-ordination processes. However, four observations go largely counter to this pattern: Table 4 reveals that officials evoking supranational allegiances tend to pay heed to signals from the political leadership (.15* and .15*), to co-ordinate with the MFA or other central co-ordinating units (.21*), and to have served for fairly long periods of time in the national central administration prior to attending EU committees (.16**). Hence, supra-nationalism seems not hampered by national co-ordination or governmental steering.

The above contradictory empirical patterns might be interpreted as follows: First, the re-socialising power of EU committees may render the co-ordination reflex weaker: Despite having written instructions at their disposal, the re-socialising power of EU committees gears actors towards evoking supranational identifications. Even more, national co-ordination is often geared towards affecting the actual decision-making behaviour of the actors, not entrusting them with particular loyalties. Reflecting this, table 4 reveals that only a marginal proportion of the variance is explained. However, table 4 reveals that the general lack of national co-ordination to some extent has independent casual effect on the enactment of supranational allegiances. Hence, the general lack of ex ante national searing of EU committee participants increases the room available for evoking supranational identities.

The ambiguous empirical patterns unveiled in table 4 render it important to include all variables from tables 2, 3 and 4 into the equation. Henceforth, table 5 reveals the relative explanatory power of (i) the EU membership versus non-membership dichotomy, (ii) of the sheer length and intensity to which national civil servants attend EU committees, and finally, (iii) of different national co-ordination mechanisms.

Table 5 The relative explanatory power of (i) the EU-membership � non-membership distinction, (ii) the length and intensity to which national officials attend EU committees, and (iii) national co-ordination. A multiple OLS regression analysis (beta).

 

Allegiances towards the EU committees attendeda

Allegiances towards the EU as a whole

�An �esprit de corps� emerges over time in the EU committees�b

(i ) EU member-states versus non-member statesc

 

.16*

 

-.01

 

-.01

(ii) Affected by the EU/EEA agreementd��.

 

-.01

 

.27**

 

-.10*

Time devoted to EU related issues generally�.

 

-.01

 

.07

 

.15*

Time devoted participating on Commission expert committees and Council working parties����..

 

 

 

 

-.06

 

 

 

 

.01

 

 

 

 

.21**

Number of EU committees attended��..

 

-.11*

 

-.05

 

.25**

Number of meetings in EU committees last year...

 

.15*

 

.07

 

.19**

Do you give oral presentations during EU committee meetings?��.

 

 

-.10*

 

 

-.01

 

 

.10*

Face-to-face contacts outside formal EU committee meetings��..

 

 

.37**

 

 

.13*

 

 

-.24**

Contacts by phone, e-mail, fax, etc. outside formal EU committee meetings�

 

 

.02

 

 

-.13*

 

 

.13*

Number of years having attended EU committees...

 

.01

 

-.09

 

-.10*

(iii) Doing clearances with other central administrative institutions before attending EU committeese�����..

 

 

 

 

-.19*

 

 

 

 

-.05

 

 

 

 

.06

Doing clearances with the MFA��������..

 

-.30**

 

-.15*

 

.04

Paying heed to signals from the political leadership nationally��.

 

 

.09

 

 

.25*

 

 

.12*

�I co-ordinate with the MFA or with other central co-ordinating units���..

 

 

.36**

 

 

.05

 

 

.02

�My �position� has been co-ordinated with all relevant ministries���...

 

 

-.14*

 

 

-.02

 

 

-.01

�I have clear instructions as to what �positions� to pursue in EU committees�

 

 

-.02

 

 

-.18*

 

 

.11*

*) p .05��������� **) p .01�������������������������� R2 = .22��������������������������� R2 = 17����������������������������� R2 = .28

a)       Dependent variables: Variable 1 and 2 applies the same value labels as table 2.

b)       Dependent variables: cf. a.

c)       Independent variable: The first variable has the following values: Sweden and Denmark (EU members) (1), Norway (EEA member (2).

d)       Independent variables: Cf. the value labels from table 3.

������ Independent variables: Cf. the value labels from table 4.

A first look at table 5 reveals that the proportion of explained variance is significantly higher than in tables 3 and 4. Moreover, table 5 uncovers that the sheer number of controlled effects that load significantly on the enactment of supranational allegiances are greatest on the intensity dimension. There are relative few co-ordination variables having a significant effect on supra-nationalism. Second, the EU membership versus non-membership turn out having a significant positive effect on the enactment of allegiances towards the EU committees attended (.16**). Hence, Danish and Swedish government officials tend to establish allegiances towards EU committees to a larger extent than the Norwegian civil servants.

Moreover, the interview data render ample support for the intensity thesis:

We talk together at the pub in the evenings. Then we get good contact. The same individuals arrive at several meetings. We call each other and send e-mails to each other. An �esprit de corps� emerges in the group� (Source: Norwegian agency official � author�s translation). �We have regular e-mail contacts. An enormous feeling of collegiality emerges amongst the Nordic colleagues. A strong sense of allegiance develops in the groups when you have participated for 5 years. It becomes your baby. It becomes so collegial� (Source: Norwegian agency official � author�s translation). �The dinners in the evenings are important arenas for talk. We spend a lot of time together� (Source: Norwegian agency official � author�s translation). The same official reports that �I travel from Norway as an �ambassador� for Norway, only to return as a representative for the EU Commission. I often feel stronger allegiance to my Swedish colleague than to my Norwegian colleagues� (author�s translation). �We have frequent contacts between the meetings, rather informal personal contacts. This result in a certain allegiance to the committee and to the individuals who attend� (Source: Swedish agency official � author�s translation).

Despite the various countervailing observations made in table 5 [5] , the vast majority of the above observations support the predicted pattern: There are a greater number of significant effects supporting the proposed pattern, and these effects are generally stronger than the deviant effects. The general observation thus holds: Supra-nationalism is strongest amongst Danish and Swedish officials, strongest amongst those officials devoting much time and energy participating on EU committees, and strongest amongst those officials whose �positions� are poorly co-ordinated and weakly instructed at the national level of governance.

However, two general countervailing tendencies stand out: First, officials evoking supranational allegiances tend to have participated for only shorter periods of time on EU committees. Yet, this observation may reflect the fact that supra-nationalism reflects less the sheer length to which national officials attend different EU committees and more the intensity of current and instant involvement (Trondal 2001).

Second, some officials evoking supranational identities also tend to participate on EU committees with a low degree of intensity. The three observations deviating most strongly from the predicted pattern will be discussed in greater detail in the following. First, the strongest deviant observation of table 5 reveals that officials having a great deal of informal face-to-face contacts with fellow committee participants outside formal committee meetings report that an �esprit de corps� emerges to a lesser extent compared to officials having fewer face-to-face contacts (-.24**). This negative relationship, however, may reflect the fact that officials having many informal contacts outside the EU committees become less strongly exposed to the �esprit de corps� emerging within each committee. The second strongest deviant finding of table 5 reveals that officials that pay heed to signals from the national political leadership tend to develop allegiances towards the EU as a whole (.25*). This may reflect the fact that, despite strong political signals, the EU committees manage to re-socialise the participants. Hence, this observation may reflect the re-socialising power of the EU committees. Finally, the strongest deviant observation of table 5 reveals that officials co-ordinating with the MFA or with other central co-ordinating units tend to evoke allegiances towards the EU committees attended (.36**). This observation might also reflect the re-socialising power of EU committees. However, one caveat is warranted. National co-ordination and national political signals are generally not geared towards entrusting the delegates with particular institutional allegiances. National instructions are more geared towards controlling and monitoring the actual moves made by each EU committee participant (cf. above). Hence, despite strong national co-ordination and governmental control, there is plenty of room left for supranational allegiances to emerge. This might be due to the re-socialising power of EU committees, triggered by the intensity to which national officials attend EU committees.

 

Conclusions

Despite providing ample evidence that membership in the EU indeed propels supranational identifications amongst EU decision-makers, the current article reveals that the EU membership versus non-membership distinction is not the only explanation of supra-nationalism, perhaps not the most adequate. Foremost, the current study highlights the re-socialising power of EU committees. Hence, when studying supra-nationalism at the level of the individual civil servants, one should pay heed to the intensity to which national civil servants attend EU committees. Finally, supra-nationalism also stems from the general lack of ex ante co-ordination mechanisms at the national level of governance.

This conclusion lends support to an institutional approach to supra-nationalism. Institutions embedding civil servants matter as far as the evocation of particular allegiances and identifications are concerned. Even more, the empirical observations support a multilevel version of institutionalism in two respects. First, accompanying multiple institutional affiliations at different levels of governance, civil servants are shown to evoke a multiple set of national, sectoral and supranational identities. Parallel to Haas (1958), intensive and sustained participation on EU committees gear national civil servants towards shifting loyalties and institutional allegiances towards the EU level. However, whereas Haas (1958) argued that supranational loyalties replace pre-existing national identifications, the current study shows that supranational identifications merely supplement pre-established national and sectoral allegiances. Second, a multilevel version of institutionalism is empirically supported in the sense that the diverse institutional allegiances evoked by the civil servants reflect the multiple institutional affiliations embedding them. Contrary to Haas (1958), supra-nationalism reflects not only the re-socialising impact of EU institutions on national decision-makers, but also particular institutional dynamics of national government institutions embedding EU decision-makers.

Endnotes



[1] Paper to be presented at ECSA�s 2001 International Conference, May 31-June 2, Madison, Wisconsin, and at the SCANCOR workshop on Transnational regulation and the transformation of states, June 22-23, Stanford University, California. The current study is financed by the ARENA-programme (The Norwegian Research Council). An earlier version of this paper was preented at the Norwegian conference in political science 10-12 January, 2001. Thanks to Morten Egeberg, Arild Farsund, Hans-Petter Graver, Leif Helland and Hans Robert Zuna for valuable comments.

[2] In the current article the conception of supra-nationalism parallels the notion of supranational identities, as defined here.

[3] The following ministries and agencies are included in this study:

Denmark: The Department of Environment and Energy, the Environmental Protection Agency, the National Forest and Nature Agency, the Department of Labour, and the National Labour Market Authority.

Norway: The Ministry of Environment, the Norwegian Pollution Agency, the Ministry of Local Government and Regional Development, and the Directorate for Labour Inspection.

Sweden: The Ministry of Environment, the Environmental Protection Agency, the Ministry of Industry, Employment and Communication [3] , and the Occupational Safety and Health Administration.

[4] One of the above observations is needless to comment because it unveils spuriousness in the OLS regression analysis presented in table 5. This relationship is marked by a parenthesis in the above table.

[5] Officials evoking supranational allegiances tend to be fairly weakly affected by the EU/EEA agreement (-.10*), attending few EU committees (-.11*), give few oral presentations during EU committee meetings (-.10*), have few informal face-to-face contacts outside formal meetings with fellow EU committee participants (-.24**), have few informal contacts by phone, e-mail or fax (-13*), and having attended EU committees for only a few years (-.10*). Second, national civil servants enacting supranational allegiances tend to pay heed to signals from the political leadership (.25* and .12*), co-ordinate with the MFA or with other central co-ordinating units (.36**), have clear instructions (.11*), and having served for fairly long periods of time in the national central administration prior to attending EU committees (.-09*).

 

References

Andersen, S. S. (2000) �EU Governance: A New Institutionalist Turn?�, unpublished paper, The ARENA programme, University of Oslo.

Beyers, J. (1998) �Where does supra-nationalism come from? The ideas floating through the working groups of the Council of the European Union�, European Integration online Papers (EIoP) 2 (9): http://eiop.or.at/eiop/texte/1998-009a.htm

Egeberg, M. (1999) �Transcending Intergovernmentalism? Identity and Role Perceptions of National Officials in EU decision-making, Journal of European Public Policy 6 (3): 456-474.

Egeberg, M. and Trondal, J. (1999) �Differentiated integration in Europe: the case of the EEA country Norway�, Journal of Common Market Studies 37 (1): 133-142.

Gulick, L. (1937) �Notes on The Theory of Organizations. With Special References to Government in the United States�, in L. Gulick and D. Urwick (eds), Papers on the Science of Administration, New York: Institute of Public Administration, Columbia University.

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

Hammond, T. H. (1990) �In Defence of Luther Gulick�s Notes on the Theory of Organization, Public Administration 68: 143-173.

Hanf, K. and Soetendorp B. (eds) (1998) Adapting to European Integration. Small States and the European Union, London: Longman.

Hooghe, L. (1999) �Supranational Activists or Intergovernmental Agents? Explaining the Orientations of senior Commission officials towards European Integration�, Comparative Political Studies 32 (4): 435-453.

Jacobsson, B. and Sundstr�m G. (2000) �Europa til vardags�, in K. St�hlberg (ed), Kontinuitet och f�rnyelse � europeisk integration och nodisk f�rvaltningsanpassning, Nord 2000:3, Nordic Council.

Jensen, C. S. (2000) �Neofunctionalist Theories and the Development of European Social and Labour Market Policy�, Journal of Common Market Studies 38 (1): 71-92.

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

Knill, C. and Lenschaw A. (2001) �Revolutionary or marginal changes in Europe? Rearranging the literature on institutional change to tackle the question of scope�, Comparative Political Studies 34 (forthcoming).

Lewis, J. (1999b) �Theories of European Integration. Rationalism and Sociological Institutionalism: Ontological Rivals�, in Methods of Community and The Permanent Representatives to the European Union, unpublished PhD dissertation, University of Wisconsin-Madison.

Lewis, J. (2000) �The methods of community in EU decision-making and administrative rivalry in the Council�s infrastructure�, Journal of European Public Policy 7 (2): 261-289.

L�greid, P. (2000) �Europeisering og forvaltningstilpasning i Norden�, in K. St�hlberg (ed), Kontinuitet och f�rnyelse � europeisk integration och nordisk f�rvaltningsanpassning. Nord 2000:3, Nordic Council.

McDonald, M. (1998) �Identities in the European Commission�, in N. Nugent (ed), At the Heart of the Union. Studies of the European Commission, Houndmills: MacMillan Press.

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

M�ny, Y., Muller P. and Quermonne J.-L. (eds) (1996) Adjusting to Europe. The impact of the European Union on national institutions and policies, London: Routledge.

Metcalfe, L. (1994) �International policy co-ordination and public management reform�, International Review of Administrative Sciences 60: 271-290.

Moravcsik, A. (1998a) The Choice for Europe. Social Purpose and State Power from Messina to Maastricht, Ithaca: Cornell University Press.

Niemann, A. (1998) �The PHARE programme and the concept of spillover: neofunctionalism in the making�, Journal of European Public Policy 5 (3): 428-446.

Pescosolildo, B. A. and Rubin B. A. (2000) �The Web of Group Affiliations Revisited: Social Life, Postmodernism, and Sociology�, American Sociological Review 65: 52-76.

Peters, B. G. (1999) Institutional Theory in Political Science. The �New Institutionalism�, London: Pinter.

Rosamond, B. (2000) Theories of European Integration, Houndmills: Macmillan Press.

Sandholtz, W. and Stone Sweet A. (eds) (1998) European Integration and Supranational Governance, New York: Oxford University Press.

Schaefer, G. F. (2000) �Linking Member State and European Administrations � The Role of Committees and Comitology�, in M. Andenas and A. T�rk (eds), Delegated Legislation and the Role of Committees in the EC, The Hague: Kluwer Law International.

Schaefer, G. F., Egeberg M., Korez S. and Trondal J. (2000) �The Experience of Member State Officials in EU Committees�, EIPASCOPE, No. 3: 29-35.

Simon, H. (1997) Administrative Behavior. Fourth Edition, New York: The Free Press.

Stubb, A. C.-G. (1996) �A Categorization of Differentiated Integration�, Journal of Common Market Studies 34 (2): 283-295.

The Prime Minister�s Office (1997) Regjeringens of departementenes arbeid med E�S-saker og enkelte WTO-saker. Oslo.

Tranholm-Mikkelsen, J. (1991) �Neo-functionalism: Obstinate or Obsolete? A Reappraisal in the Light of the New Dynamism of the EC�, Millennium: Journal of International Studies 20(1): 1-22.

Trondal, J. (2001) Administrative Integration across Levels of Governance. Integration through Participation in EU Committees. Unpublished PhD dissertation, University of Oslo (forthcoming).

Trondal, J and Veggeland F. (2000) �Access, voice and loyalty. The representation of national civil servants in EU Committees�, ARENA working paper (8).

Wessels, W. and Rometsch D. (1996) �Conclusion: European Union and national institutions�, in D. Rometsch and W. Wessels (eds), The European Union and member states. Towards institutional fusion?, Manchester: Manchester University Press.

Yin, K. R. (1993) Applications of case study research, Newbury Park: SAGE.