ARENA Working Papers
WP 98/20

 

 


Consociationalists or Weberians? Top Commission Officials on Nationality



Liesbeth Hooghe
Department of Political Science, University of Toronto



 


Abstract

This paper examines how senior permanent officials in the European Commission (directors-general and directors) conceive of the role of nationality in their organization. Do they support a weberian ideal-typical bureaucratic organization, where merit shapes personnel selection and task organization, or do they prefer a consociational type, in which nationalities are represented in organization and policy making? I explain variation in weberian and consociational orientations, using 105 mail questionnaires collected between July 1995 and May 1997. In explaining variation, I contrast socialization factors and factors related to the professional utility function of officials. I find that utility packs far more power than socialization. Utility is most strongly influenced by nationality: support for consociational principles is highest among nationals with large personnel quotas in the Commission and nationals with strong supportive networks in Brussels. In an administration where nationality is a powerful principle of personnel organization, officials with the “right citizenship” have compelling incentives to reinforce its role. Professional utility is also a function of one's position in the work environment: officials in positions of weak regulatory autonomy or dealing with quality of life issues are more likely to be consociational. Socialization is weak, though prior experience as national civil servant reduces consociationalism and prior Commission cabinet experience increases it.

--------------------------

This article examines how senior permanent officials in the European Commission conceive of the role of nationality in their organization.* How do they deal with the fact that their colleagues have different national backgrounds and that their main clients are national governments with divergent interests? Do senior Commission officials consider themselves guardians of common European interests that transcend national particularities? Or do they conceive of themselves as the agents of national stakeholders in European policies? There are two approaches to deal with multi-nationality rooted in distinct conceptions of political-bureaucratic organization. One is the weberian ideal-type, according to which merit shapes personnel selection and task organization, and rational criteria guide policy making. Commission officials' attitudes and behavior are expected to reflect the general European public interest rather than national interests. The second model is consociational, in which political-bureaucratic organization and policy making represent the diversity of the polity. Proportionality determines how the common realm is carved up among constituent units, while consensus seeking and recognition of each group's vital interests determines policy making (for an application to the EU, see Taylor 1991, 1997). In such a system, Commission officials reproduce national diversity in the EU, ultimately by serving as agents of their nation.

This tension between merit and representation is present in every international organization, but it is particularly acute in the European Commission. First of all, a growing proportion of authoritative decisions affecting the lives of Europeans are taken at the level of the European Union, and the European Commission is the executive-bureaucratic core of this transnational system. It wields influence over a wide range of domestic policies, and it has exclusive competence to initiate and draft legislation (Nugent 1995; Marks, Hooghe, Blank 1996; Wallace and Wallace 1996). What public interest these Commission officials represent matters a great deal. Furthermore, disagreement about the proper balance between weberian and consociational principles of organization in the Commission is not new. The Commission was originally conceived as an unambiguously weberian organization. The first president of the European Commission, Jean Monnet, crafted a small, highly professional team of permanent officials, who were to embody a higher European interest, formulate common problems and solutions to the College of Commissioners, and persuade on their behalf national representatives to adopt supranational arrangements. However, national governments have been reluctant to let control over European governance slip. This is reflected in the growth of the Council machinery to counterbalance the Commission's organizational resources and in the expansion of comitology in the 1970s and 1980s to curb the Commission's executive autonomy. It underlies attempts to control the Commission from the inside through national quotas for recruitment, influence on the appointment of top officials (most particularly through parachutage), by questioning tenure for Commission officials and by encouraging their substitution with seconded national officials.

Against the backdrop of this ongoing tug of war between institutions, where do top officials stand on merit and national representation in the Commission? Should the Commission embed national diversity into its administrative and policy organization in consociational fashion? Or should the Commission be insulated from contending national interests, and model itself on a rational-legal conception of bureaucracy?

The following section describes weberian and consociational principles in the Commission's formal and informal rules and top officials' orientations. I then develop hypotheses to explain variation in orientations. One line of theorizing explores to what extent consociational or weberian orientations may be the result of socialization in particular institutional environments. A second approach examines how orientations may be understood as a function of the opportunities inculcated in a particular institutional and situational context. I test these hypotheses against data collected between July 1995 and May 1997 from interviews with 137 senior Commission officials of A1 or A2 grade (director-generals, deputy director-generals, directors and principal advisors) and mail questionnaires from 105 of these people. I find that officials' orientations are most strongly associated with how consociational and weberian principles affect their professional career. More generally, the findings raise doubts about the power of socialization in a relatively porous institutional setting as the Commission.


The Commission and Nationality

The organization of the Commission reflects both consociational and weberian principles. For recruitment and promotion at top levels, consociational norms and practices prevail. Senior posts are divided among nationalities according to quotas that roughly reflect the distribution of votes in the Council of Ministers. Furthermore, personnel regulations make it possible to attract external candidates for top bureaucratic positions if no suitable internal candidate can be found. Nearly half of senior appointments are recruited through parachutage, that is, they are appointed in A1 or A2 positions from outside the Commission; the other half are career Commission officials promoted from in-house middle management. The influential role of national governments in the selection of these “parachutists” is well-documented (Ross 1995; Nugent 1995; Page 1997). So even though, in principle, top officials are appointed on the basis of merit, in practice, they need the right nationality and (preferably) the blessing of their national government.

Consociational principles are also present in the administrative organization of the Commission. As a matter of principle, the most senior civil servant of each directorate-general (DG), the equivalent of a ministerial department, must have a different nationality from the responsible Commissioner. [1] Officials in adjacent positions in the chain of command are generally not of the same nationality. And all directorate-generals, directorates (large subdivisions), units (next subdivision) and task forces (temporary units) have a policy of maximizing “geographical diversity” among their personnel, which means that they aim to have a variety of nationalities and in addition often a balance between north and south. [2]

The explicit recognition of national diversity in the administrative organization of the Commission effectively limits national bias in policy making. First, it reinforces the general norm inherited from Monnet that national favoritism should be frowned upon (Coombes 1970; Egeberg 1995; Ludlow 1991; Page 1997). Monnet wanted the Commission to be the “platonic embodiment of Communitarian spirit, with gallic élan, self confidence and expertise” (Quoted in Church and Phinnemore 1994: 270). Furthermore, multinational balance in units and across hierarchical levels makes it very difficult for individual officials to bestow favors on their nationality unseen. It also inhibits colonization of particular units by a nationality or group of nationalities. This finds support in the available empirical research on nationality in the Commission. Page has examined whether national colonization takes place in the Commission by comparing the actual distribution of nationalities in DGs with the expected distribution in non-colonized DGs. He has detected no general evidence of colonization, at least not at the level of directorate-generals (Page 1997). Focusing on national favoritism by individual officials, Egeberg has found limited traces in less than one quarter of Commission units. He has concluded that, “what has emerged seems to be more than just a secretariat to the Council, or a neutral broker. ... intentionally or unintentionally shaped, the services seem to have achieved some autonomy for promoting common European interests” (Egeberg 1995: 28).

A major exception to the weakness of consociational principles in Commission policy making consists of the cabinets. Each political Commissioner—two each for the five largest member states and one for the other member states—has a group of 5-10 political aides. Their role is to reconcile general European interests with the party-political and national interests of their Commissioner. Cabinet members are almost always of the same nationality as the Commissioner. Because of their unique consociational characteristics Commissioners and their cabinets have been described as “national enclaves” in the Commission (Michelmann 1978; see also Egeberg 1995; Nugent 1995; Peterson 1997; Ross 1995; Spence 1994).

How does this mixture of rational-legal norms for policy making and consociational principles of organization affect Commission officials' perceptions? Table I shows responses on two questions. The first question taps into the national colonization issue, and the second indicates national favoritism by individual officials. Thirty-nine percent of Commission officials think that national colonization is a problem. Individual national favoritism is perceived as somewhat less problematic: close to 30% believe that too many Commission officials let their nationality interfere with their policy decisions. All in all, a substantial minority disapproves of the role for nationality in the Commission, and between 9 and 12% does so quite unequivocally. These officials seek to strengthen rational-legal principles of organization in the Commission. However, the majority is satisfied with the current balance between consociational and rational-legal principles, and between 23 and 28% candidly reject concerns about national capture. As a group, then, top Commission officials are bent to the consociational status quo end of the scale.

Table I: Top Officials and Nationality in The Commission

Items*
Yes(4) Yes, but No, but No(1) Mean**
It hurts the Commission's legitimacy that certain DGs tend to be dominated by particular nationalities, such as agriculture by the French, competition by the Germans, regional policy by the Spanish, environment by the north ... 1211.4% 2927.6% 3432.4% 2826.7% 2.248
Too many Commission civil servants let their nationality interfere in their personal judgments. 109.5% 2120.0% 4845.7% 2321.9% 2.186

* n=105. Absolute figures do not add up to 105 because of abstentions. Idem for percentages.

** Responses range from 1 to 4. Neutral position would be 2.5; an average below 2.5 indicates disagreement with the statement.


Conceptualizing Top Officials’ Orientations to Nationality

Imagining a European public interest requires a leap beyond the real world of diverse nationalities. Conversely, accommodating national diversity in consociational fashion presupposes the relaxation of modern weberian conceptions of bureaucracy. Why do officials display one or the other disposition?

My point of departure is that the tension between consociational and weberian criteria shapes top officials' views about nationality. There are two basic approaches to understanding contention among top officials. The first, rooted in a sociological model of belief system constraint, emphasizes socialization. Individuals may spend time in institutional environments supporting weberian or consociational principles, so that their experiences accumulate over time to coagulate in congruent beliefs (Converse 1964; Mughan et al. 1997; Rohrschneider 1994, 1996; Searing 1969, 1985; Verba 1965). A second line of theorizing entails a more voluntaristic, incentive-induced, and more situational understanding of elite beliefs. Two assumptions underlie an incentive-induced logic of adaptation to institutional constraints. First, institutional rules have calculable consequences for strategies, relative power and objectives of actors (Hall and Taylor 1996; North 1990; Ostrom 1990, 1991; Steinmo, Thelen and Longstreth 1991). Institutions limit, constrain and condition choice--not only over time, but here and now. Secondly, actors are rational, in the sense that they usually want to make the best of a “constrained” situation and that they usually want rules that increase their chances to achieve their goals. The definition of rationality used here is not the narrow economic definition, but a less restrictive rationality principle (Ostrom 1990, 1991; Searing 1991). Institutions provide incentives for actors to choose to desire, think or act in particular ways. Consociational and rational-legal criteria of organization have calculable professional consequences for top officials. Rational-legal criteria insulate them from political/ national machinations. Consociational criteria expose them to political discretion, increasing the likelihood that nationality trumps merit in career and policy decisions. As rational actors Commission officials should want to optimize their professional opportunity structure. A logic of incentive-induced choice seeks to specify the conditions under which consociational and weberian principles of institutional design may affect top officials' views.


Hypotheses

Socialization

Prior Transnational Socialization. Officials' views on the consociational/ weberian principles of bureaucratic organization may result from prior transnational experience. Students living abroad are part of cosmopolitan communities that function according to distinct transnational norms and rules. They often return home with practices or beliefs that do not sit well with local ways. For similar reasons, officials who previously worked for international organizations are likely to be more open to conceptions of a public interest above and beyond national concerns. In short, the transnational socialization hypothesis predicts that officials who have studied or worked abroad are less likely to be consociational than those exclusively bred in their home country.

Prior National Socialization. A previous career in a national administration may influence officials' orientations in two ways. A simple version links prior national civil service to resistance to consociational principles. Most national bureaucracies in Western Europe are strongly built on weberian principles where professionalism, merit and objectivity outweigh particularistic connections and partisan judgments. National civil servants have been trained to develop a sense of public service. On the basis of the socialization logic they are likely to extrapolate their national experiences to the Commission. One may therefore hypothesize that former national civil servants support weberian rather than consociational principles. A sophisticated version expects that former national officials extrapolate the particular norms and practices in their home administration to the Commission. Not every national administration is equally weberian; consociational practices have been extensive in Austria, Belgium, Italy, Luxembourg, and to some extent the Netherlands. So one may hypothesize that former national bureaucrats from consociational administrations are more comfortable with consociationalism in the Commission than their colleagues from weberian administrations.

Commission Socialization. Under the auspices of Jean Monnet, the Commission was originally conceived as a professional team of permanent officials without ties to member states. In this weberian institution, Commission officials were expected to espouse a sense of mission to transcend interstate relations and construct “an ever closer Union”. To the extent that these values are embedded in the Commission as institution, one may hypothesize that the longer officials work in the Commission, the more they should internalize these institutional norms and reject consociational principles. This socialization thesis is echoed in anthropological studies of the Commission (Duchêne 1994, on Jean Monnet; Abélès, Bellier, McDonald 1993; Bellier 1995; McDonald 1997; Shore and Black 1992).

Commission Cabinet Experience. A final socialization hypothesis links cabinet experience to consociational principles. In his study of the Delors period, George Ross describes the role of cabinets in balancing national interests, party-political priorities and European political goals with technocratic policy making (Ross 1995). One may hypothesize that officials with cabinet experience are more consociational.

Professional Opportunity Structure

I examine how three situational settings may affect professional opportunities: rules for recruitment, rules and practices related to particular positions and tasks in the Commission, and characteristics of the nationality to which officials belong.

National Control over Recruitment. One hypothesis links national control over recruitment to officials' orientations. In the US political system, control over the bureaucracy is ensured through the spoils system combined with mandatory approval of top federal administrators by the Senate. National governments in the European Union do not have formal control over senior Commission officials. However, they are usually consulted informally on top appointments. This national leverage is strongest for appointments through parachutage, the recruitment of externals for A1 or A2 positions. It seems plausible that national governments will lobby for candidates favorable to consociational principles and that officials who owe their appointment to their national government will be more inclined to be responsive to national concerns. Whether on the basis of selection processes or power relations, one may hypothesize that parachuted officials should be more consociational than non-parachuted officials.

Positional Interest. How to strengthen one's professional position is likely to differ from position to position. Officials in areas of strong EU competencies are not much dependent on national governments' consent to get things done. But they need ongoing regulatory or financial autonomy, and these resources are better guaranteed in a weberian administration. In contrast, officials with limited regulatory responsibilities often find other attributes more essential than competencies or funds: access to information, mediation skills, capacity to use persuasion, and credibility to exert social pressure on national governments. For them, close interaction with governments is often beneficial. So one may hypothesize that Commission officials will be less consociational to the extent that they are in positions of greater policy autonomy.

Substantive Policy interest. A second hypothesis concerning the Commission's internal work environment links substantive policy tasks with officials' views on nationality. There are several reasons why quality of life issues may induce officials more readily to favor consociationalism than established low politics (internal market, industrial policies, redistributive policies) or high politics (external affairs). Competencies on quality of life issues are often shared among levels of governments, which creates strong incentives for cooperation. Furthermore, national variation in preferences and institutional arrangements is significant so that attention to national sensitivities is likely to produce better results than insisting on a common European interest. From their side, national advocates of environmental, cultural or consumer issues have strong incentives to use the EU arena to achieve policy goals that are difficult to obtain domestically, and this should facilitate close links between national and European policy makers. (On EU environmental policy, Sbragia 1996; on EU social regulation, Eichener 1992.) One may therefore hypothesize that officials dealing with quality of life issues have incentives to favor consociational principles.

Nationality. In an administration where nationality is a powerful principle of organization, certain nationalities are associated with better career opportunities or greater effective weight in policy making. For officials with less advantageous nationalities the rational strategy is to pursue a Commission administration that is relatively insulated from national influences. One may hypothesize that nationalities that do not benefit from current consociational rules should prefer a Commission based on weberian principles.

The utility of nationality for professional life is influenced by two factors. First, a promotion system that gives priority to nationality over merit imposes severe career constraints on officials from small countries. This is a direct consequence of the proportionate size of national quotas. Small nationalities like the Danish for example may expect between 7 and 10 top positions, while the four largest nationalities claim between 27 and 32. With an average annual turnover rate of fewer than 10 top Commission positions (and fewer still in years of enlargement), it can take several years, sometimes over half a decade, before one position opens up for small nationalities. Under a merit system Danish officials could compete annually. One may therefore hypothesize that officials from smaller countries are less likely to be in favor of the consociational status quo.

A second factor concerns the effectiveness of national networks. Some nationalities have a strong reputation of “club-ness”, which I define as a set of formal and informal networks within which members tend to act in concert. A variety of resources may contribute to clubness. One resource, often mentioned in anthropological research on the European Union, is national socio-cultural cohesion. Cut off from their home environment, individuals with strong national identities tend to socialize with compatriots. These informal national networks on golf courses, in bars or literary evenings easily become invaluable venues for professional contacts among compatriots (Abélès, Bellier and McDonald 1993). Clubness may also be a by-product of organizational and financial resources. The sheer concentration of administrative-financial resources in the larger national communities should enable them to better monitor and lobby Commission personnel policy. This argument is similar to that made by research on the presence, cohesion and effectiveness of state delegations to the US congress, which has found strong associations with population size, size of state bureaucracy and professionalism (Morrisroe 1998). Finally, clubness may be the result of a deliberate policy by national governments or other national actors to strengthen networks among expatriates in Brussels. Whatever the base, strong clubness turns nationality into an asset for officials competing for professional advancement. One may therefore hypothesize that officials from strongly networked nationalities are most likely to favor consociational principles.


Data

Orientations to National Responsiveness. To measure top officials' stance to nationality, I combine the two items of Table I (with reversed coding) in an index for National Responsiveness. Item one focuses on individual national favoritism; item two asks officials to take a stance on national colonization. [3] Values range between 1 (weberian) and 4 (consociational). The mean is 2.783 out of 4, with a standard deviation of 0.795. I refer the reader to the appendix for greater detail.


Independent Variables

Prior Transnational Socialization. I use a dummy variable for Transnational Experience. Commission officials who studied abroad or have worked for international organizations are assigned a value of one; 43 % entered the Commission with transnational experience.

Prior National Socialization. For the simple state socialization hypothesis, a dummy for State Experience takes on the value of one when officials spent time in the national state sector. Fifty-eight percent worked for the state, where they spent on average 10 years.

Reliable measurements of the consociational/ weberian character of bureaucracies are hard to come by. Page has compared bureaucracies through four characteristics—cohesion, autonomy of political control, caste character, and non-permeability for interests, which can be conceived as dimensions for categorizing bureaucratic traditions (Page 1995). Using this research I divide the fifteen member states in three categories ranging from weak over medium to strong weberian administrations. I then create two interaction terms, Strong Weberian and Weak Weberian, which combine State Experience on the one hand and dummies for strong and weak weberian categories on the other hand. The reference group consists thus of former state officials from medium weberian administrations.

Commission Socialization. I use Length of Service, that is, the number of years served in the Commission until the interview. A top official has spent on average 18 years in the Commission, ranging from a few months to 38 years.

Cabinet Socialization. Thirty-five percent of top officials served in a cabinet, and nearly half of those for more than one Commission term. I construct a dummy Cabinet, with a value of one for officials with cabinet experience.

Positional Interest. To assess the extent to which Commission officials wield autonomous power, I combine formal and reputational measures of Commission power in a composite index. For the former, I rely on figures compiled by Page on secondary legislative activity by the Commission (Page 1997). In addition, I use a reputational question posed to the 137 top officials to name the three or four most powerful DGs at the time of the interview. The measure for PowerDG is a ranking from 1(weak DG) to 8 (powerful DG).

Substantive Policy Interest. I use a dummy Quality DG to distinguish between officials dealing with quality of life issues (value of 1) and those that do not. Nearly 23% of the officials work on issues of gender, equal opportunity, third world solidarity, environment, consumer rights, participation, citizenship rights, or social and educational rights.

National Control over Recruitment. A dummy variable Parachutage has a value equal to one if an official was appointed from outside into a top position, and a value of zero for an official promoted from inside the Commission. Of the sample, 42% are parachutists.

Nationality. To test the nationality utility hypothesis I construct two variables. One taps into the argument that the size of national quota affects opportunities for career mobility. For National Quota I allocate to each official the number of votes in the Council of Ministers for his country of origin, which range between two and ten. This is a conservative measure because in reality small nationalities usually receive fewer jobs than suggested by the Council voting key.

It is more difficult to measure the effectiveness of national networks. I draw from descriptive accounts (Christoph 1993; Cini 1996; Dutriaux 1994; Grant 1994), anthropological studies (Abélès, Bellier and McDonald 1993; Bellier 1995; McDonald 1997; Ross 1995), primary sources and my own interviews to assess the effect of cultural cohesion, financial and organizational resources and intentional policy. I divide the nationalities in three categories for National Clubness ranging from weak (1) to strong (3) clubs.


Examining the evidence

A First Cut: More Opportunity than Socialization

The sources of officials' divergent views on nationality are to be found in a mixture of socialization factors and factors characterizing their professional opportunity structure, in which opportunity packs considerably more power than socialization. Four sets of factors account for 42% of the variance. By far the most powerful association is with the pair testing the nationality utility hypothesis, which jointly explain 24% of the variance in the bivariate association (Table II, column 3: r2 for nationality) and also dominate the multivariate analysis (model 3 and 4). Officials competing within the limited promotion opportunities of small national quotas are more resentful of nationality as organizational criterion in the Commission (r2=0.086). And even more powerfully, officials are much less likely to embrace consociational principles when they belong to a national group that supports its members ineffectively (National Clubness: r2=0.209). A second set of variables pertains to the opportunity structure within the Commission. Officials from powerful DGs (external relations, competition, and agriculture) are most weberian, and officials dealing with quality of life issues are most consociational. Both associations are about equally powerful in the simple regression (column 3), but the former weakens considerably with controls (Models 3 and 4). Among the socialization variables, state socialization beats all other variables with 13% of the variance. Former national officials usually want a more weberian Commission, though those from cohesive, autonomous, impermeable administrations are more likely to champion consociational principles. I will discuss this apparently counter-intuitive finding below. Finally, top officials with Commission cabinet experience are significantly less worried about nationality.

Model 1 and 2 test the effect of socialization and professional opportunity separately. The professional opportunity models explains 37% of the variance, which is considerably more than the 17% for the socialization model. Top officials know that the Commission's approach to nationality profoundly shapes their professional opportunities. They make reasoned assessments of how more consociationalism may affect them, and they form orientations accordingly.


A Second Cut: Socialization Outflanked by Opportunity

What is the rationale behind this constellation? Are all officials equally receptive to professional opportunity incentives or are the views of some shaped by experiences inculcated over years? And is nationality for all such a determining factor—be it as parameter in the competition for professional success or as a proxy for the lasting impact of particular political-bureaucratic norms and practices? To examine these questions I carve up the group of 105 officials according to recruitment channel to a top career. This juxtaposes “consociational products”, who achieved top positions through the consociational channel of parachutage, and “weberian recruits”, who reached the top through promotion through the ranks. EU studies are divided about desires and deeds of parachutists and internal recruits. With respect to the level of support for consociational principles, two arguments are made. For some, parachutists are Trojan horses sent by member states to undermine the European Commission's autonomy from within; they are likely to favor consociationalism. Others point out that many parachutists resemble “the economics professor who happened to find favor with the government before last of a particular member state” (Page, 1997:139); parachutists may want an administration according to rational-legal principles. With respect to this question, the results of this study are unambiguous: as the bottom row of Table III shows, parachutists are significantly less keen on consociational principles than internal recruits.

There are also contrasting expectations with respect to the sources of variation in consociational support. On the one hand, parachutists are relatively unaffected by the fact that professional opportunities vary by nationality. Contrary to internal recruits, parachutists were catapulted into the top administration over and above national competition; one may therefore expect their views to be shaped by transnational and, especially, Commission socialization. However, others do not anticipate a great influence of socialization factors, first and foremost because socialization requires common history, and history is not what most parachutists have in common with the Commission. They believe that neither socialization nor nationality utility, but opportunity and constraint in the parachutists' work environment shape their orientations. Table II presents the analysis for all officials and Table III for parachutists and internal recruits. The remainder of the paper examines key results for each variable.

Transnational Socialization. The transnational socialization hypothesis predicts that officials who studied or worked abroad should be less responsive to nationality. The bivariate analysis gives moderate support, but when one controls for other, more powerful factors, the variable drops out. Transnational experience varies by nationality, and this effect is picked up by nationality-specific variables.State Socialization. The simple state socialization hypothesis, which predicts that former state officials favor weberian principles, finds support. Though the effect is modest in the simple regression, it grows for the total sample (model 3 and 4). This effect is amplified for ex-civil servants from weak, incohesive, highly permeable and politicized administrations (model 3 and 4). The negative effect of state socialization is reversed to a positive one for officials from strong, cohesive, impermeable administrations with limited politicization (model 3 and 4). So the sophisticated hypothesis, which argues that officials should extrapolate norms and practices from their home administrations to the European Commission, is soundly rejected; instead, the relationship is there, but the sign is opposite to the one predicted. Officials from consociational as well as those from weberian heritage tend to support principles that are the opposite to their national experiences. The overall explanatory power of the three state variables in the simple regression is larger than the sum of its parts, which suggests that one rather than three separate logics seems to drive this factor (r2=0.131). These conclusions are generally applicable to parachutists and internal recruits, though a different mix of state variables does the explanatory lifting. The theoretical implication of these findings is that socialization, at least for elite actors, is not a mechanistic process. Though they internalize norms and practices from previous settings, they re-evaluate them before applying them to a new institutional context. Hence the bottom line is the same for all: prior state service makes them less likely to support consociational principles. But, all other things equal, they add lessons drawn from their particular administrative experiences. Officials from consociational national administrations know full well that consociationalism has often led to severely restricted policy capacity, low status and alienation for civil servants, and so they prefer a Commission built on weberian principles. Officials from weberian national administrations realize that weberian traditions rest on conditions absent in the European Commission—homogeneous cultural and educational background of civil servants, and the presence of a unitary political principal—, and so they appear willing to endorse the consociational status quo. The logic at work is not merely socialization; lesson drawing complements it. Commission Socialization. Proponents of the strong socializing capacity of the Commission will find little solace in this study. The variable drops out in all multivariate analyses. To the extent that length of service in the Commission matters, it encourages officials to be responsive to nationality (Table II: r=0.153). Most officials have learned that it is difficult to get things done if one ignores national sensitivities. Yet the bottom line is that the Commission finds it hard to mold the orientations of its top employees to its institutional self-interest. Elsewhere, I have argued that the Commission is a greenhouse neither for supranationalism nor for a particular ideology (Hooghe forthcoming, 1998). This paper argues that it is even less a breeding place for a European public interest that distances itself from national sensitivities.

Cabinet. The hypothesis that officials with cabinet experience are more responsive to nationality finds support, though it is only moderately strong in the multivariate analysis (model 3 and 4: significant at .1 level). The findings for the subgroups show why the effect is deceptively modest. Among parachutists, views on nationality do not differ between ex-cabinet members and others; all are reluctant supporters of consociational principles. But among those who were promoted from the Commission's middle management, ex-cabinet members are far more accommodating to nationality than those without cabinet experience.

A major reason why the association between cabinet experience and views on nationality is high for internal recruits and not for parachuted officials has to do with the differential centrality of cabinet service for career advancement. For ambitious middle-management officials, cabinet service is a crucial stepping stone to a high-flying career, because that gives them a chance to get noticed by Commissioners. About 40% of internal recruits have taken this route. [4] Commissioners tend to select compatriots who understand the interests of the Commissioner's country of origin and are willing to balance European concerns with national sensitivities. Once immersed in cabinet politics, officials become further attuned to diverse national interests. A mixture of self-selection, selection (by Commissioners), and learning, within severe nationality constraints on career advancement, explains why ex-cabinet members are more open to consociational principles: their score is 2.98 out of a maximum of 4 instead of an average 2.78 for all officials and 2.63 for parachutists. Parachuted officials make very different calculations. By the time they join a cabinet—and 30% do—they have already landed a top job. A cabinet posting is not a stepping stone to a more senior position, but a political interlude in an already successful, primarily administrative-managerial career.

All in all, I find the explanatory power of socialization limited. State socialization is largely a misnomer for a much more cognitive process of lesson drawing. Socialization factors are conspiciously absent in explaining where parachutists stand. Even for internal recruits, the one strong socialization factor, cabinet experience, is perhaps better interpreted as a disguised professional opportunity variable. The following paragraphs turn to professional opportunity factors.

Positional and Substantive Policy Interest. I consider two ways in which professional opportunities may be influenced by officials' work environment. The positional interest hypothesis predicts that the more officials have institutionally entrenched policy autonomy, the less they will support consociational arrangements. This hypothesis finds considerable support in simple (r2=0.055) and multiple regressions (model 3 and 4) and the association is particularly strong among parachutists. The substantive interest hypothesis suggests that officials dealing with quality of life issues are more likely to favor consociational principles. There is strong support in the simple (r2=.049) and multiple (model 3 and 4) regressions and across the two subgroups. There is no evidence that socialization is at work rather than situational response to a given incentive structure. To test for socialization I count for each official the number of years spent in PowerDGs and QualityDGs respectively and I correlate these figures to the dependent variable. The associations are modest for power DGs (r=-0.167, p=0.08) and low for DGs dealing with quality of life issues (r=0.140, p=0.153). Even more tellingly, officials currently working for powerDGs do not become less consociational as they serve longer (r=0.049, p=0.815); similarly, among those working on quality of life issues, veterans are not more consociational than newly appointed officials (r=0.121, p=0.508). The defining feature is where one works in the Commission, not since when.

Parachutage. Parachutage gives national governments the illusion of consociational control, but it tends to strengthen the camp of those defending the European public interest. Contrary to expectations, parachutage is not positively associated with support for consociational principles. The simple regression suggests the opposite association (Table II), though the variable drops out in the presence of controls. This finding goes against the dominant view among EU scholars, but it is not surprising for students of consociationalism. It is fairly easy to restrict positional access to individuals with the appropriate characteristic—party membership, ethnicity, language, religion, or national citizenship; it is much more difficult to control views and decisions of appointees on an ongoing basis. This is particularly so when appointees are protected by restrictive tenure regulations, as is the case for senior Commission officials.

While this helps us understand why parachutists are not necessarily proponents of consociationalism, it does not explain why they should be less enthusiastic than internal recruits. This result is partly spurious: there are fewer ex-cabinet members and more individuals from nationalities with small national quota and weak clubness among parachutists. But a conceptual argument for their reluctance is suggested by Aberbach, Rockman and Putnam' s seminal work on politics and bureaucracy. They conclude that bureaucrats in highly politicized systems, in particular Italian and Belgian particracies, are most deeply alienated from politics (Putnam 1977, 1976; Aberbach, Rockman and Putnam 1981). Prior dependence upon non-merit career criteria breeds distancing from such criteria after promotion. It is plausible that the same processes influence parachuted Commission officials.

Nationality. The hypothesis that officials' views on consociational principles in the Commission are critically influenced by the utility of national citizenship for career purposes finds overwhelming support. The two variables are highly significant in simple and multivariate models (models 3 and 4). The subpopulations clarify the dynamics. National quotas constrain aspiring top officials from smaller countries more than their colleagues from larger member states. The “sense of having the wrong nationality” is particularly salient among internal recruits because their chances for promotion are most sharply circumscribed by the national quota system. Parachutists, who are usually asked to apply for a position reserved for a particular nationality, are much less sensitive. These conjectures are borne out by the very strong positive association between national quota and consociational principles in models 7-8 (internal recruits) and the non-significant association in models 5-6 (parachutists).

The most powerful association is with national clubness. National networks shape one's career opportunities, but they also matter in subtle ways for people at the top. Abélès, Bellier and McDonald emphasize the social support provided by national networks, but they also recognize more instrumental functions, such as exchange of information, contacts with influential compatriots in and outside the Commission, and political opinionating.

“Each nationality has its club, its network, its association of European officials, its “church”, and they are especially frequented by those officials who are most destabilized by the multinational work environment. These happen to be more often Irish or Danish than German or Italian. Not all officials have a need to come home. Membership of the Irish club provides gossip, makes it possible to keep up with local news. Equally so, the Dutch, the Danes ... try to find in Brussels the pubs where they can bump up against one another – without having to make prior arrangements: a national habit. The Portuguese club groups ambassadorial diplomats and permanent representatives to NATO and the European Community. With its thematic dinners spiced up by reputable speakers, it performs a social and intellectual function. The French participate in political associations or, for the products of the Ecole Nationale d'Administration, in “old boys networks”. The Spanish form a small colony, but the nocturnal social life has had to give way to the exigencies of the [Brussels] climate and the work rhythm in the Commission. The British, members of a club in London, do not see the need to belong to a club in Brussels.” (my translation from French, pp.25-26)

National networks do not only have social and professional use for officials; they are also transmission belts between the Commission and its interlocutors. Many officials play a role as “points of access” for governments, firms and other interest groups of their country of citizenship (Egeberg 1995). In a polity where successful policy making often depends on the quality of intelligence, officials with weak national networks are at a disadvantage.


Conclusion

This article examines how elite officials in the European Commission conceive of the role of nationality in the Commission. At stake are two distinct conceptions of political-bureaucratic organization. According to the weberian ideal-type, merit rather than nationality shapes personnel selection and task organization, and officials' attitudes should reflect the general European public interest. In a consociational model, the diversity of the European polity should be reproduced in the Commission's organization, and officials should reflect the various national views. This paper argues that, as a group, top officials are bent to the consociational status quo end of the scale. Yet there is considerable variation in their orientations to nationality. So why are some officials consociationalist while others call for a more autonomous European public interest?

The search for sources of variation is set up as a contest between socialization factors and factors related to the professional opportunity structure of officials, in which opportunity packs far more power than socialization. Nationality shapes professional opportunities profoundly and so it is no surprise that nationality factors are highly significant. The greatest support for consociational principles is to be found among officials from large nationalities and from nationalities with strong supportive networks in Brussels.Yet professional opportunities are also affected by one's position in the work environment. Being in a position with weak regulatory autonomy or dealing with quality of life issues induce officials to support a more consociational, responsive approach to nationality. To some extent officials' views are also shaped by experiences inculcated over time: prior experience as national civil servant reduces consociationalism and prior cabinet experience increases it.

I have found not one, but at least two different types of officials: parachutists and internal recruits. Different causal processes underlie variation in their orientations. As one moves from internal recruits to parachutists nationality factors increasingly give way to influences having to do with life and work in the Commission.

Internal recruits. Nationality is a powerful predictor of internal recruits' position on the consociational-weberian continuum. Their chances for promotion are most sharply circumscribed by the national quota system and by their access to strong national networks. However, internal recruits' beliefs are also influenced by incentives in-house, though less strongly. The most significant determinants are how much institutional power they master and what substantive policy area they oversee (quality of life or other), but cabinet socialization also molds their views. Internal recruits seem torn between the Commission's consociational personnel policy—where nationality sets the pace for career advancement—and the Commission's relatively weberian policy organization and style—where functional and positional incentives shape preferences. In their eyes, the Commission appears highly permeable to external national influences for career matters, and relatively bounded for policy making.

Parachutists. The world looks different from the perspective of parachutists. They get their cues to a larger extent from inside the Commission: their positional and substantive policy interests influence their views deeply. Certainly, parachutists' orientations are still powerfully influenced by whether they belong to a strong or weak national support network, but parachutists do not feel the competitive pressure associated with severely limited national quotas. The Commission appears to them as relatively bounded—though not invulnerable to external influences.

A broader conceptual-methodological lesson can be learned from this study. EU politics has traditionally been understood as a function of nationality. But little effort has been devoted to unpacking the causal processes behind the proper nouns. I demonstrate that the critical role of nationality in shaping officials' views on the Commission's organization does not mean that officials reflect divergent national interests on the status of the Commission in the European polity. Nor does it evince that officials simply bow to the authorities that appoint them: parachutists to national governments and internal recruits to the Commission. Rather, nationality figures prominently for instrumental reasons: support for a consociational Commission depends on whether one's nationality is associated with better career opportunities and greater effective weight in policy making. In an administration where nationality is a powerful principle of organization, officials with the “right” citizenship have compelling incentives to reinforce the role of nationality.

Many studies of European integration have assumed that the Commission is intent on substituting diverse national concerns with a uniform European interest. This paper disconfirms this assumption for the Commission's elite officials. There is a surprisingly great acceptance that Europe's diversity should be explicitly recognized in the Commission. In the words of one official: “I like my service to be a microcosm of the Community. I like my colleagues to reflect the diversity within the Community. There is a certain mystery as to how people with such different backgrounds can work together.” National diversity will be a fundamental feature of the European polity for a long time to come.


Appendix

Data. I interviewed 137 out of 200 top Commission officials of A1 and A2 status between July 1995 and February 1997, and received from them 106 mail-back questionnaires with behavioral questions and 32 statements measuring orientations on political and social life (n=106). I use 105 for data analysis here; one questionnaire was excluded because a comparison with corresponding interview excerpts raised doubts about the validity of some responses. A comparison between these samples on key characteristics (position, age, nationality, gender, education, prior career, seniority, Commission cabinet experience, parachutage, nationality) reveals no bias. Testing sample bias with respect to the population is more difficult as the Commission does not publish socio-demographic data for its top officials. However, I test sample bias for nationality by using as yardstick common Commission practice to seek a “geographical balance” in top appointments, that is, reflecting the distribution of seats in the Council of Ministers. French, British and to some extent Italian and Dutch officials appear over-represented in my questionnaire sample, while nationals from the second (Greek, Portuguese and Spanish) and third enlargement (Austrian, Finnish and Swedish) are underrepresented. However, the chi-square statistic falls short of rejecting the null-hypothesis that distribution in sample and population are the same (alpha=.436).

Dependent variable. Additive index of the two items in Table I, divided by two (Cronbach's alpha of 0.63). Principal component factor analysis identifies a single dimension, with eigenvalue of 1.46 and 73.1% of variance explained.

Transnational Experience. Dummy takes on value of one if international education or work experience abroad. Source: Biographical data from: The European Companion. London: DPR Publishing, 1992, 1994; Euro's Who's Who. Brussels: Editions Delta, 1991; interviews by author.

State Service. Dummy for national service. These concern positions in the executive branch of the state and hierarchically subordinate to central government: civil servants in line ministries, diplomats (excluding EU postings), and government ministers (but not national parliamentarians). For public officials with some autonomy from central authorities (courts, central bank, parliament, public companies, local government) or in positions with a strong European component (European desks in Foreign Affairs or near the head of government), I allocate half of the time to state career. Source: biographical data.

Strong/ Weak Weberian. Two interaction dummies between State Service and dummies for strong/weak weberian bureaucratic tradition. I apply four categories for comparing bureaucracies developed by Page to classify traditions into strong/ medium/ weak weberian bureaucratic tradition. I use country assessments by Page where possible. Strong cohesion: France, Ireland, UK. Autonomy of political control: much (Denmark, Ireland, Sweden, UK), some (France, Germany, the Netherlands, Portugal, Spain), little (others). Caste-character: France, Germany, UK. Non-permeability for outside interests: France, Ireland, Portugal, Spain, UK. So I divide national bureaucracies in three categories: strong weberian (France, Ireland, Germany), medium weberian (Denmark, Germany, Netherlands, Portugal, Spain, Sweden), weak weberian (Austria, Belgium/ Lux, Finland, Greece, Italy). Source: Page (1995).

Length of (Commission) Service. Years in Commission service. Source: biographical data.

Cabinet. A dummy with value one for those with cabinet experience. Source: biographical data.

PowerDG. A composite index of formal and reputational measures. For the formal indicator, Page measures two types of secondary legislative activity by the Commission: regulations, directives and decisions that require Council approval, and those that do not. As there are no official statistics on legislative output per DG, Page and his collaborator White used keywords (author; form; year; subject) to scan the Justis CD-Rom for legislation over the period 1980-94 (over 30,000 pieces), and allocated output to the DG considered to be the most plausible author. I did a manual recount for 1980-94 for some policy areas, and arrived at a comparable breakdown. Amendments to Page's data pertain to DGs created since 1994. The proportion of autonomous Commission output to total output ranges between 10% (budget) and more than 90% (competition). Values: 1= up to 20%; 2=21-40%; 3=41-60%; 4=61-80%; 5=above 80%. Sources: Page (1996); European Commission. N.d. Directory of EU legislation in Force until Dec 1994. For the reputational indicator, I use a question posed to the 137 top officials in which they name the three or four most powerful DGs or services in the Commission at the time of the interview. DGs with a high reputation (mentioned by 50 % or more) obtain a value of 2, those with medium reputation (mentioned by 5-49 %) 1, and the remainder 0. I then add scores for the two indicators to create PowerDG.

Quality DG. A dummy taking a value of one for officials working in DG V, VIII, X, XI, XXII, XXIV. Source: biographical data.

Parachutage. A dummy taking a value of one for officials parachuted from outside the Commission in A1 or A2 positions. Source: biographical data.

National Quota: The number of votes in the Council of Ministers for officials' country of origin, ranging between two and ten.

National Clubness. An index composed of assessments along three dimensions. Strong cultural cohesion is characteristic of the Austrian, Dutch, Irish, Portuguese and the three Scandinavian nationalities (Abélès, Bellier, McDonald 1993). Especially the Irish have a strong reputation in Brussels for social networking. The fact that Belgian officials live in their country rather than in an expatriate community may impede club formation. Secondly, the organizational and financial resources of the French, British, German, Spanish and to some extent Italian communities outweigh those of any other nationality. Thirdly, clubness may be forged by intentional policy. One indicator is direct national intervention, usually through the government; this is particularly outspoken for the French, British, German and Spanish. French and British governments/ civil services closely monitor personnel policy in the Commission and consider postings in Brussels as an integral part of the training for their best and brightest (Lequesne 1993). For the French, this is part of a more general policy to organize French citizens scattered over European and international institutions. German officials do not feel so much the influence of their capital, but of their national political parties, which in line with domestic practice divide senior German posts in Brussels among themselves. Partisanship is likely to emerge as an effective channel for Austrians and perhaps the Finnish as well. Spanish governments have a reputation for pushing their nationals hard in career matters, though this approach is sometimes weakened by sharp partisan conflicts. Pro-active governmental or partisan lobbying sits uncomfortably with the strongly merit-based traditions of the Dutch, Scandinavians and to a lesser extent the Portuguese. Finally, deliberate policy has traditionally been ineffective or unwanted by officials of the three remaining nationalities: no government tops the Greek government's reputation of ineffective performance in general and in personnel lobbying in particular; many Belgian and Italian officials have distanced themselves from their clientelistic home base (interviews). Another indicator of intentional policy to forge clubness is the extent to which Commission cabinets give priority to the career concerns of their compatriots. For senior appointments Commissioners of the involved nationality are usually consulted, but some take such consultation more seriously than others. As pointer of the importance attached to personnel issues, I have coded number and rank of those responsible for personnel in each Commissioner's cabinet under the Santer Commission. German, Swedish, and British cabinets devote most resources; followed by French, Italian, Irish, Portuguese and Spanish cabinets; further by Belgian, Dutch and Luxembourg cabinets; and finally by Austrian, Danish, Greek and Finnish cabinets (Source: American Chamber of Commerce in Belgium. 1997. EU Information Handbook. Brussels.) On the basis of these three streams of evidence, I divide the nationalities in three categories: weak clubness (Belgo-Luxembourgers, Greeks, Italians); medium clubness (Dutch, Scandinavians, Portuguese, Spanish); and strong clubness (Austrians, British, French, German and Irish).


References

Abélès, Marc, Irène Bellier, Maryon McDonald. 1993. Approche anthropologique de la Commission européenne. Brussels: report for the European Commission (unpublished).

Aberbach, Joel, Bert Rockman and Robert Putnam. eds. 1981. Bureaucrats and Politicians in Western Democracies. Cambridge: Harvard University Press.

Bellier, Irène, 1995. “Une culture de la Commission européenne? De la rencontre des cultures et du multilinguisme des fonctionnaires.” In: Politiques publiques en Europe. Yves Mény, Pierre Muller, J.-L. Quermonne, eds. Paris: L'Harmattan, 49-60.

Christoph, J.B. 1993. “The Effects of Britons in Brussels: The European Community and the Culture of Whitehall.” Governance 6, 4:518-37.

Church, Clive and David Phinnemore. 1994. European Union and European Community. A Handbook and Commentary on the Post-Maastricht Treaties. New York: Harvester Wheatsheaf.

Cini, Michelle. 1996. The European Commission: Leadership, Organization and Culture in the Commission. Manchester: Manchester University Press.

Converse, Philip E. 1964. “The Nature of Belief Systems in Mass Publics.” In: Ideology and Discontent. David E.Apter ed. New York: Free Press of Glencoe.

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

Duchêne, François. 1994. Jean Monnet: The First Statesman of Interdependence. New York: Norton.

Eichener, Volker. 1992. Social Dumping or Innovative Regulation? Processes and Outcomes of European Decision-Making in the Sector of Health and Safety at Work Harmonization, EUI Working Paper.

Egeberg, Morten. 1994. “Bridging the Gap between Theory and Practice: The Case of Administrative Policy.” Governance 7:3-98.

Egeberg, Morten. 1995. “Organization and Nationality in the European Commission Services.” Arena Working Paper, Oslo, 13/95.

Grant, Charles. 1994. Inside the House that Jacques Built. London: Nicholas Brealey.

Hall, Peter and Rosemary Taylor. 1996. Political Science and the Three New Institutionalisms. Paper presented at Max Planck Institut für Gesellschaftsforschung, Köln, 96/6.

Hooghe, Liesbet. “Supranational Activists or Intergovernmental Agents? Explaining Political Orientations of Senior Commission Officials to European Integration.” forthcoming in Comparative Political Studies 32:4.

Hooghe, Liesbet. 1998. “Euro-Socialists or Euro-Marketeers? Explaining contending Orientations to European Capitalism among Senior Commission Officials.” Paper presented at the Annual Meetings of the American Political Science Association, Boston, 1998.

Lequesne, Christian. 1993. Paris-Bruxelles: Comment ce fait la politique européenne de la France. Paris: Presses universitaires de la Fondation nationale des sciences politiques.

Ludlow, Peter. 1991. “The European Commission” In: The New European Community: Decision-Making and Institutional Change. Robert Keohane and Stanley Hoffmann eds. Boulder: Westview Press.

Marks, Gary, Liesbet Hooghe and Kermit Blank. 1996. “European Integration since the 1980s. State-Centric Versus Multi-Level Governance.” Journal of Common Market Studies 34: 343-78.

McDonald, Maryon. 1997. “Identities in the European Commission.” In: At the Heart of the Union: Studies of the European Commission. Neill Nugent ed. London: MacMillan, 27-48.

Michelmann, Hans. 1978. Organizational Effectiveness in a Multinational Bureaucracy: The Case of the Commission of the European Communities. Westmead: Saxon.

Morrisroe, Darby A. 1998. “The State of `State Offices' in Washington: A Critical Assessment.” Paper presented at the Annual Meeting of the American Political Science Association, Boston.

Mughan, Anthony, Janet Box-Steffensmeier and Roger Scully. 1997. “Mapping Legislative Socialization.” European Journal of Political Research 32:93-106.

Nordlinger, Eric. 1981. On the Autonomy of the Democratic State. Cambridge: Harvard University Press.

North, Douglass C. 1990. Institutions, Institutional Change, and Economic Performance, Cambridge: Cambridge University Press.

Nugent, Neil. 1995.The Government and Politics of the European Union. London: MacMillan.

Ostrom, Elinor. 1990. Governing the Commons. Cambridge: Cambridge University Press.

Ostrom, Elinor. 1991. “Rational Choice Theory and Institutional Analysis”, American Political Science Review, Vol.85, No.1, pp.237-43.

Page, Edward. 1997. People Who Run Europe. Oxford: Oxford University Press.

Page, Edward. 1995. “Administering Europe.” In: Governing the New Europe. Jack Hayward and Edward Page eds. Durham: Duke University Press, 257-285.

Peterson, John. 1997. “The Santer Commission: Pillarized, Nationalized, or Just Normalized?” Paper prepared for presentation at the biennial conference of the European Community Studies Association Centre, Seattle, 29 May-1 June.

Putnam, Robert. 1977. “Elite Transformation in Advanced Industrial Societies. An Empirical Assessment of the Theory of Technocracy.” Comparative Political Studies 10, 3: 383-411.

Putnam, Robert. 1976. The Comparative Study of Political Elites. Englewood Cliffs, NJ: Prentice-Hall.

Rohrschneider, Robert. 1996. “Cultural Transmission versus Perceptions of the Economy. The Sources of Political Elites' Economic Values in the United Germany.” Comparative Political Studies 29, 1:78-104.

Rohrschneider, Robert. 1994. “Report from the Laboratory: The Influence of Institutions on Political Elites' Democratic Values in Germany.” American Political Science Review 88, 4:927-41.

Ross, George. 1995. Jacques Delors and European Integration. Oxford: Oxford University Press.

Sbragia, Alberta. 1996. "Environmental Policy: The 'Push-Pull' of Policy Making." In: Policy Making in the European Union, Helen Wallace and William Wallace eds. Oxford: Oxford University Press, 235-55.

Searing, Donald. 1991. “Roles, Rules and Rationality in the New Institutionalism.” American Political Science Review 32:47-68.

Searing, Donald. 1985. “A Theory of Political Socialization: Institutional Support and Deradicalization in Britain.” British Journal of Political Science 16:341-376.

Searing, Donald. 1969. “The Comparative Study of Elite Socialization.” Comparative Political Studies 1: 471-500.

Steinmo, Sven, Kathleen Thelen, Frank Longstreth. eds. 1991. Structuring Politics. Historical Institutionalism in Comparative Analysis, Cambridge University Press.

Taylor, Paul. 1991. “The European Community and the State: Assumptions, Theories and Propositions.” Review of International Studies 17: 109-25.

Taylor, Paul. 1997. The European Union in the 1990s. Oxford: Oxford University Press.

Verba, Sidney. 1965. “Conclusion: Comparative Political Culture.” In: Political Culture and Political Development. Lucian Pye and Sidney Verba eds. Princeton: Princeton University Press.

Wallace, Helen, and William Wallace eds. Policy-Making in the European Union. Oxford: Oxford University Press.

Table II: Multivariate Analysis--Explaining Orientations to Consociationalism

Corre-lation (r ) Simple regression(r2) Model 1 Model 2 Model 3 Model 4
Transnational Experience -.188* .036* -.113(.155)   -.056(.135) -
State Socialization

State service

Weak Weberian Tradition

Strong Weberian Tradition

-.165*

-.217**

.210**

.131***

.027*

.047**

.044**

-.327**(.235)

-.047(.257)

.325***(.234)

  -.232*(.206)

.003(.226)

.184(.225)

-.228**

(.157)

-

.182*(.182)

Length of Service in the Commission .153 .023 .098(.009)   -.100(.011) -
Cabinet Experience .183* .034* .164*(.160)   .150*(.139) .135*(.132)
Power DG -.235** .055**   -.150*(.035) -.160*(.035) -.154*(.034)
Quality of Life DG .221** .049**   .294***(.162) .297***(.163) .300***(.158)
Parachutage -.164* .027*   -.098(.192) -.098(.192) -
Nationality

National Quota

National Clubness

.293**

.457***

.240***

.086**

.209***

  .234**(.025)

.401***(.081)

.163(.028)

.343***(.096)

.156*(.026)

.365***(.008)

All variables

R2 Adj. R2

    .169.118 .372.340 .423.354 .417.373
Consociational principles(mean on a scale from 1-4)   2.78

Note: Multivariate linear regression (constant included in equation, pairwise deletion of missing values.

Standardized coefficients (betas); standard errors in brackets. Horizontal double line separates socialization and opportunity logic.

*significant at 0.10 level; ** significant at 0.05 level; *** significant at 0.01 level (one-tailed; but two-tailed for correlations).

n.s.= bivariate correlation not significant at 0.10 level (one-tailed). A dash indicates that a variable is dropped from the multivariate analysis according to pre-set criteria (p> .10).

Table III: Multivariate Analysis--Explaining Orientations to Consociationalism

Parachutists Parachutists Internal Recruits Internal Recruits
  Model 5 Model 6 Model 7 Model 8
Transnational Experience n.s. n.s. -.035(.212) -
State socialization

State service

Weak Weberian Tradition

Strong Weberian Tradition

-.026(.259)

n.s

.179(.250)

-

n.s.

-

-.159(.302)

-.207*(.489)

.139(.359)

-

-.272**(.368)

-

Length of Service in the Commission n.s. n.s. n.s. n.s.
Cabinet Experience n.s. n.s. .231**(.178) .245**(.171)
Power DG -.264*(.048) -.255*(.048) n.s. n.s.
Quality of Life DG .279**(.210) .253*(.208) .268**(.248) .283***(.227)
Nationality

National Quota

National Clubness

n.s.

.378**(.146)

n.s.

.481***(.118)

.303**(.040)

.230(.130)

.349***(.036)

.248**(.108)

R2 Adj. R2 .474.405

n=44

.454.414

n=44

.455.371

n=61

.444.393

n=61

Consociational principles(mean on a scale from 1-4) 2.63a 2.89a

Note: Multivariate linear regression (constant included in equation, pairwise deletion of missing values).

Standardized coefficients (betas); standard errors in brackets. Horizontal double line separates socialization and opportunity logic.

*significant at 0.10 level; ** significant at 0.05 level; *** significant at 0.01 level (one-tailed).

n.s.= bivariate correlation not significant at 0.10 level (one-tailed). A dash indicates that a variable is dropped from the multivariate analysis according to pre-set criteria (p> .10).


Footnotes

* This project depended on the generous cooperation of 137 senior Commission officials. The Catholic University of Brussels provided hospitality during interviewing, and the Robert Schuman Centre (EUI, Florence) gave me the opportunity to work on the project as Jean Monnet Fellow (1996-97). This article is part of a larger project supported by the department of Political Science (Toronto), and the Canadian Social Science and Humanities Research Council (grant SSHRC Research No. 72005976, Fund No.410185). An earlier version was presented at the University of Twente (July 1998), and at a conference on “Public Policy and Administration at the Turn of the Century”, organized by the IPSA Study Group on the Structure and Organization of Government, in Oxford, Lady Margaret Hall, (July 1998). I thank participants for comments and suggestions, with special thanks to Gary Marks and Bert Rockman.

[1] Exceptions are rare. The two instances during my field research concerned two Spaniards on Latin-America and the Mediterranean (DG Ib) and two British on transport (DG VII). Several interviewees expressed concern about the former, but the situation was remedied when the director-general retired in 1997. They perceived the latter as accidental and unproblematic.

[2] Administrative units that cater for a particular country or group of countries are most vulnerable to national bias. However, few units are area-specific, the main ones being in regional policy and in services dealing with third countries.

[3] These items are randomly distributed on a list with 32 items.

[4] Senior officials consider cabinet service as the most effective route to higher administrative echelons. Yet non-parachutists have two other important venues to the top: support from one's national home base (national government, party connections, national administration), or recognition of one's policy expertise or managerial excellence.






[Date of publication in the ARENA Working Paper series: 15.11.1998]