Resumé

The main purpose of this article is to refine, modify, and elaborate some central propositions and assumptions in the scholarly debate on corporatism or corporate pluralism. The empirical base is a data archive containing information on practically all interest groups in Denmark. Hypotheses are formulated and tested concerning variations in direct representation and participation of interest groups in public policy-making across (a) types of interest groups, (b) organizational resources, (c) organizational structures, and (d) political issue areas. Contrary to prevailing propositions in the literature on corporatism, the Danish case shows that corporate structures and practices go hand in hand with a system of interest groups characterized by myriads of interest groups, and an overall, rather decentralized structure.

Twenty years ago Samuel Elders veld (1958) characterized the study of interest groups and their political activities by means of a threefold classification based on the scope of the studies concerned: (1) Studies of single interest groups, (2) studies of interest groups as they operate on a single arena, and (3) studies of groups occupied with a particular law, issue area, or policy conflict. Until quite recently this characterization may be said to have been valid - at least in the Danish case. However, the last few years have witnessed not only alterations in the scope of the studies of interest groups activities: concomitantly, new approaches and key-concepts have been applied to these studies. Most conspicuously, the pluralist, laissezfaire inspired thinking of Bentley (1908), Truman (1951), and Latham (1952) has been vigorously attacked and criticized by protagonists of corporatist views. Although models of corporatism may be appropriate, and in fact have been demonstrated to be so, in the study of various

---

* This article is a revised version of a paper presented at the Fifth Nordic Political Science Meeting, Bergen, August 27-30, 1978.

aspects of functional interest representation, there seems to exist a pronounced
need for demarcating the scope of the validity of the new models
and concepts (Buksti and Johansen 1977a).

This article basically rests on the assumption that further elaboration of the model of corporatism would benefit significantly 'from systematic empirical research. Accordingly, an empirical analysis of the contacts between interest groups and public bureaucracy in Denmark is presented here. The empirical basis is deliberately broad. It covers practically all the constituent units in the overall system of interest groups; it takes in various arenas for organizational participation in public policy-making; and it concerns itself with a. multitude of political issue areas. In this way, it should be possible to overcome some of the deficiencies inherent in the older, narrow research tradition mentioned by Eldersveld. It should also be possible to capture variations in direct organizational participation in public policy-making across types of interest groups, forms of participation, andcontents of policy. In short, this design appears to be appropriate if the intention of demarcating the scope of validity of the corporatist model is to be fulfilled.

Students of corporatism usually ignore organizational resources as a relevant set of factors for understanding the relationships between interest groups and the state. This neglect is probably due to the intransigent rejection of the pluralist models and concepts in which the questions of resources and power often are thoroughly treated. In this article it is argued that certain amounts and types of organizational resources and structures are indispensable for effective actions by the interest groups in the corporate channel: they thus represent a key to the understanding of the capabilities and actual operations of interest groups in corporatist decision-making.

1. Conceptual Framework

The growing and somewhat fashionable literature on corporatism is marked by a bewildering conceptual variety and diversity, to a certain extent indicated by various prefixes like 'societal', 'liberal', 'neo', 'quasi', etc.

As far as the basic units of analysis are concerned, some definitions emphasize properties related to single interest groups or to the entire system of interest groups. Schmitter's well-known definition (1974, 93) belongs to this category. Bureaucratized, highly centralized peak organizations are also often asserted to be inextricably linked with corporate

structures and processes. Another tradition focuses on the linking structures or 'modalities' of interchange between interest groups and the state (Heisler 1974). The descriptions of corporatist features in the Nordic political systems seem particularly to stress modes of interchange, like the 'remiss' system or the system of public committees (Meijer 1969; Moren 1974; Peterson 1977; Jarlov et al. 1976; Johansen and Kristensen 1978a). Finally, a third group of authors base their definitions and conceptualizations on characteristics of the state, especially the performance of certain state functions like incomes policy or economic policy (Lehmbruch 1977; Panitch 1977; and to a certain extent Winkler 1976).

Regarding the preferred analytical dimensions, some adhere to structural or institutional mappings, while others are occupied with the processual aspects or behavioral consequences. These behavioral consequences are sometimes treated within a framework of class struggle. Thus, it is asserted that a function of suppressing the working class is performed through the corporate channel (Panitch 1977; Jessop 1978). We do not intend to cover all these aspects and analytic dimensions. On the contrary, only some structural features describing certain aspects of the interactions between the public bureaucracy and interest groups are to be discussed. This, however, does not prevent us from attempting to give some sort of meaning and content to the concept of corporatism.

Following Johan P. Olsen (1978), corporatism is conceived primarily as a special form of coordination between state authorities, particularly the public bureaucracy, and interest groups. The differentia specifica of the corporate form of coordination is the direct representation and participation of interest groups in public policy-making. This conceptualization hinges on one of the most crucial tenets in the literature on corporatism, namely the proposition on the integration or penetration of interest groups into the very apparatus of the state: that is, interest groups are not considered as standing outside, competing for access to government. Their access has become institutionalized, and administered as a matter of right by the public authorities concerned. Direct participation in public policy-making is not limited to preparatory or consultative tasks. Interest groups are also deeply and directly involved in the administration and implementation of public policies.

Direct organizational participation in public decision-making may be performed through a variety of forms: informal deliberations between representatives of interest groups and officials from the public bureaucracy; summit meetings between group leaders and leaders of cabinets; incorporation into the system of public committees, boards, commissions,

agencies. Importance is thus attached to the directness of group participation, not to the formality of this participation. Finally, it should be noticed that the corporate type of interest representation has to be dominant and salient if a given system is to be characterized as corporatist. A few isolated cases of corporatist-like features do not suffice.

As a means of understanding the variations in direct organizational participation in government, a cost-benefit perspective has been proposed (see, especially Olsen 1978; also Dahlerup et al. 1975; Heisler 1974). The benefits, the individual pressure group being the unit of analysis, may be summarized as influence, institutionalized access and public recognition, efficiency and role differentiation. The costs encompass phenomena such as loss of organizational freedom and autonomy, responsibility and moderation of demands, goal displacement, and grass root rebellion.

An application of this cost-benefit perspective implies at least two problems. First, the costs may be over-estimated due to the actual lack of alternative forms of co-ordination. In a way there is no room for a rational calculation of costs and benefits related to various forms of coordination and interest representation: the corporate one is of exceptional importance, and the organizations are in fact forced to resort to direct participation if they want to wield influence over public policies. Second, costs and benefits, looked at from the state or state authorities, are neglected. Technical information and legitimation are often asserted to constitute the main benefits to the state. Using the same terminology, the cost-side - in the form of problems of state management, co-ordination and steering - seems to be under-estimated.

Yet this approach still offers a preliminary clue to the fact that direct participation in government is not equally attractive or possible for all types of organizations, and not equally relevant for all types of issue areas. Bearing these comments in mind, we can now turn to the task of examining variations in direct group participation in government across type of organization, organizational resources and structures, and political issue areas.

2. The Data Archive on Danish Interest Organizations

The analysis is based on a data archive set up in 1976-77, containing information on all Danish national interest organizations or organizations that strive to be national in scope. In terms of the conceptual framework, the data archive is designed to make possible overall analyses of public policy formation in general and patterns of interaction between organizations

and public authorities in particular. Consequently, it is based upon a rather large data set. Using the obvious advantage of studying a relatively small, and in many politically relevant respects, homogeneous country such as Denmark, a comprehensive questionnaire was administered to all the relevant organizations. The response rate was nearly 85 percent. The collected data contain information on organization size; year of formation; economic, technical, and administrative resources; organizational structure; type, scope and frequency of contacts to public authorities; goals and objectives of the organization; inter- and intra-organizational linkages; international relations and activity, especially towards the European Communities (Buksti 1979). Besides these questionnaire data, the archive consists of statutes, annual reports, and other publications and communications from the organizations involved.

The criteria for demarcating the relevant cases were that an interest organization in this context has to be (i.e., clearly defined membership codified in a set of rules or statutes), voluntary (i.e., the members are as a common rule not fully employed in the organization or earn their primary income via the membership), and national association of individuals, firms, institutions., associations, organizations, etc. In spite of some obvious problems, this definition has at least two advantages. It is close to the major part of definitions in the research on interest groups (Wilson 1973:31-32), and it has proved to be functional and operational in a data-collection phase (Buksti and Johansen 1977b).

In addition, the following supplementary criteria have been applied: (1) In certain areas of society - due, for instance, to historical traditions, geographical conditions, administrative structures and especially the general economic, social or political development of the society - there are no national organizations. In these fields the existing regional organizations are included in the analysis, because otherwise one would have a wrong impression of the degree of organization in the society as a whole. (2) Public or semi-public agencies, institutions or organizations are excluded. (3) Purely commercial agencies or private firms were left out. (4) Political parties are not included. On the basis of these criteria, the actual number of organizations included in the data archive is 1953, including 125 regional organizations. To obtain an overview of the great variety of organizations involved, they were classified in the following 19 categories of organizations (Buksti and Johansen 1977b; 1977c).

1. Professional organizations: Blue-collar workers (N = 56)

4. Employers' organizations (117)
5. Agriculture, forestry, fishing (119)
6. Industry and handicraft (162)
7. Tråde and commerce (213)

8. Banking, insurance, transport, press (71)
9. Housing organizations (6)
10. Consumers' organizations (18)

11. Organizations for local authorities (18)
12. Organizations for schools and education (43)
13. Organizations for art and culture (92)
14. Scientific societies (173)

15. Productivity and technology (30)

16. Social welfare and humanitarian organizations (137)
17. Organizations for youth, sport, and recreation (138)
18. Religious organizations (39)

19. Organizations with nationalistic aims and organizations for international
cooperation and friendship (89)

This categorization is based on several different criteria combined in different ways in the individual categories. The latter are in some cases rather complex and in some ways overlapping. They represent a rough, but operational classification of the system of interest organizations. The main division goes between categories 1-8 on the one hand and the remainder on the other. The first group is primarily based on the profession and education of the members, and the organizations represent certain functional interests. The other group includes organizations which (1) are not primarily based on the profession of the members (except the scientific societies); which (2) in principle address themselves to the general population. Some of the latter are primarily oriented towards economic or production questions (9-11), others are only concerned with technical information and scientific discussions (14-15), while the rest try to promote certain ideas, values or ideologies (12-13, 16-19). Of all the organizations included in the archive some 60 per cent are oriented primarily towards economic matters, 25 per cent towards social ideas and values, while 15 per cent are primarily concerned with technical information and scientific problems.

3. Variations in Organizational Participation in Government

3.1. 'Professional' and 'Promotional' Interest Grups

It is a commonplace observation in the literature on both corporatism and interest groups that some interest groups participate more and more deeply in public policy-making than others. Some authors a priori delimit corporate decision-making to the incorporation of certain types of interest groups, e.g. economic producer groups or the organized expressions of socio-economic functional interests (Panitch 1977; Lembruch 1977), or to interest groups based on the division of labour in society (Schmitter 1977). Others point to a distinction between professional or sectoral groups and promotional groups. The first category comprises groups which are related to the central and relatively permanent roles of citizens, i.e. occupational and educational roles. Groups of this kind are, accordingly, characterized by the fact that they address themselves to specific, welldefined and delimited segments or sectors in society. The second type has in contrast the whole population as a potential 'constituency', and the groups concerned are occupied with the promotion of certain ideas, values, beliefs, ideologies. The former type is expected (and can be demonstrated) to be more heavily involved in interplay with public authorities than the latter (Finer 1966; 1973; Beer 1958; Olsen 1978). A third way of differentiating interest groups proceeds from the specific system of incentives that a given group applies towards potential members, actual members and activists, in order to maintain and strengthen its position. Some groups are primarily marked by their use of material sanctions (both positive and negative), while others rest upon ideological or solidary incentives (Wilson 1973).

It appears to be fairly reasonable to hypothesize that promotional groups or groups based on ideological or solidary incentives are less involved in corporate decision-making. Indeed, the costs in the form of loss of organizational freedom and autonomy, prevention from claiming pure ideological positions, moderation and adjustment are perceptibly at stake for these groups. On the other hand, the 'benefit-side¹ seems to be more obvious for professional or sectoral groups. They may be able to convert direct engagement in public policy-making into resources to be used for the well-being of the organization. In a certain sense they have to let themselves be incorporated into the corporate structure, because the activities of the state so directly and significantly interfere in their field of interest and operation and, concomitantly, in the lives of their members.

It is, however, important to note that this hypothesis asserts only that

professional groups are on the whole more involved in corporate structures and processes than are promotional groups. It does not preclude cases of participation by promotional groups. Cases like this may in fact occur - though at a significantly lower level and more intermittently.

If one looks at the sheer number of interest groups which in one way or
another have been in contact with public authorities during the period
under investigation (1975), the figure turns out to be 53 per cent.1 Of the

professional groups 52 per cent report contact, while the figure for the promotional groups is 54 per cent. Superficially, this result is indeed surprising. It may, however, hide enormous differences in the intensity and scope of the actual interaction.

In order to capture differences like this, the emphasis in what follows is put solely on contacts with the public bureaucracy. Moreover, three dependent variables measuring various aspects of direct participation are used. First, the frequency of participation is measured. Some groups participate regularly or even daily, while others do so only occasionally. Second, the scope of the contacts is assessed. Some organizations may participate in public policy-making covering a broad array of ministries, while the activities of other groups may be restricted to one ministry only. Third, the form of participation is tentatively measured in terms of the number of public committees, commissions, boards or agencies in which the organizations are represented. As is argued elsewhere, the system of public committees adjacent to the public, central administration seems to be one of the most vital, if not the single most important, institutional arena in which the interplay between interest groups and the state takes place (Johansen and Kristensen 1978b).

A preliminary test of the hypothesis concerning professional and promotional groups is presented in Table 1. The table clearly shows that promotional groups, when compared to professional groups, tend to participate only occasionally; they direct their activities towards one ministry only; and they are represented in only a few public committees. Yet it should not be overlooked that a remarkable number of these promotional groups actually participate in government, and Xhatsome of them do so to a considerable extent.

As far as the differences among the professional groups themselves are concerned, a certain pattern seems to show up. Blue-Collar Unions and the organizations of Business, Agriculture, and other economic organizations are on the whole more heavily involved than White-Collar Unions and Employers' organizations. This, however, may be caused by differences in the spread of contacts among organizations within the same type. The more a given kind of contact, let us say daily contact, is spread across different organizations within the same type, the more contacts for that category of organizations, ceteris paribus. Conversely, the more the contacts are concentrated in one or few organizations, the fewer organizations with contacts of that type appear in the table. This quite obviously is the case for the employers' organizations, where all contacts at the highest level, in terms of frequency, scope and form go through only one organization-The

Danish Employers' Confederation. The degree of centralization within organizational sets seems to determine the access structure or the pattern of direct participation, but not necessarily the total level of participation.

Moreover, the kind of display in Table 1 conceals the fact that even organizations characterized by the same degree of frequency, scope, and form of contact have different types and amounts of resources to bring into play. In order to achieve a fuller understanding of how deeply those organizations already incorporated in fact participate in corporate decision-making, it is necessary to refine considerably the simple typology pui: forward above. It is particularly important to explore further the extent to which organizational resources and structures determine the pattern of interchange between interest groups and the public bureaucracy. This is the topic of the next section.

3.2. Organizational Resources

By resources of an organization we mean everything that the particular organization can mobilize for realization of its goals. Obviously, these very broadly defined resources may vary considerably depending on the specific issue concerned, the contextual situation, and the character of the goals involved. Some organizations may function only as service arrangements to the members, while others may partly take care of certain activities principally affecting their members in the political decisionmaking arena. Consequently, some organizations are passive in relation to public authorities, because of their limited goals, while others may be very active in promoting or defending their interests in relation to these authorities. Different goals may demand different resources.

Further, the value of the different resources controlled by a specific organization will depend on the needs and interests of other political actors in the political system. Control of resources of direct importance to political decision-making or to the implementation of political decisions may be of decisive significance to the organization concerned. Control of such resources would be a strong reason for acceptance of the organization by the political apparatus, and it would be the basis for closer and closer contacts of the organization with the political decision-making structures. Obviously, organizational resources may have decisive consequences for the participation of interest organizations in government and for their behaviour in the corporate channel.

In a study of factors affecting interest group strategies in the United
Kingdom, Grant (1977) makes a distinction between insider groups and

outsider groups on the basis of the relationship between groups and
government. Elaborating on this terminology, the following typology of
interest organizations emerges:

Pure outsiders are either organizations with very few resources in general or with resources that are not important to the government and public authorities. They cannot gain attention and be accepted as legitimate spokesmen that have to be listened to by political decision-makers. They lack adequate resources, and they are - so to speak - outsiders by necessity (Grant 1977). They may, however, also be organizations with limited goals that concern members only, and consequently are normally not likely to seek contact with public authorities.

Deliberate outsiders are organizations that control some resources of importance to the government and public authorities in certain political situations. But they are careful not to become too entangled with the political decision-making system, because they may wish to challenge accepted authorities, institutions, or ideas. They are outsiders by choice (Grant 1977). On the one hand, deliberate outsiders can be former insiders' that have decided not to participate in the corporate structures any longer and not to continue regular contacts with public authorities for political or ideological reasons. This, for instance, has been the case for the National Union of Danish Students. On the other hand, deliberate outsiders may be ad hoc organizations, grass roots movements, or movements formed across the established political segments in society or around new policy areas, like for instance the Organization for Information about Atomic Power (00 A) in the field of nuclear energy. They try to realize their goals through extra-parliamentary activities, primarily by influencing public opinion. Consequently, the capacity for mobilizing rank and file members for collective action is a very important resource for such organizations.

Impotent insiders are organizations that wish to enter corporate structures or to come into regular contact with public authorities, and that are accepted by these authorities. The participation of these organizations is encouraged by the government or a particular public authority. But the organizations concerned lack adequate resources and the capability to make independent moves, though they are consulted in all questions of concern to them. This, for instance, has been the case for the Danish fishermen's organizations. They face more problems than they are able to cope with, and thus, most problems are postponed and thereby intensified in some way. They are not able to act according to environmental changes; and their impotence makes it very difficult for them to operate in the corporate structure (Buksti 1976).

Finally,pure insiders are organizations deeply involved in the political and administrative processes. They participate very intensely in the corporate structures, and they actively try to expand those structures. They control great economic resources and a high degree of technical and administrative expertise, which seem to be the most relevant resources for organizations participating intensely in corporate structures. These organizations are often characterized by a high degree of bureaucratization and professionalization of leadership. In Denmark, the existence of pure insider organizations has been quite obvious, especially in the field of agriculture and the labor market, but lately also in other areas, as, for example, with the National Association of Local Authorities.

The two constituting dimensions of this typology, however, may be mixed, and the four above-mentioned types of organizations should be considered as no more than ideal types. Further, according to our conceptual framework, it is primarily the insider organizations that will be reviewed, while in this context outsider organizations are merely of theoretical interest.

Looking at the four types of organizations, there seem to be some basic resources of general significance for the degree of organizational participation in the political decision-making process. These resources are: number of members, economic strength, and size of the secretariat or permanent staff of the organizations. However, number of members and economic strength are very difficult to operationalize for systematic comparative analysis. Organizations may have different kinds of members - e.g., individuals, factories, institutions, other organizations - while economic resources may be difficult to use because of a reluctance to make such information public. Further, economic resources may derive from many different sources like the members, 'commercial' activities,

special funds and even from state subsidies, and this may affect the
behaviour of the organizations concerned.

Finally, this typology of organizations demonstrates the importance of the size of the secretariat or the bureaucracy of organizations in relation to their ability to join corporate structures. It reflects the ability and the willingness of organizations to use resources in their interactions with public authorities. The existence of a permanent staff is fundamental to the capacity of organizations to act and perform according to environmental changes (Terreberry 1968:607), and it is essential to interaction with the political apparatus as long as both government and organization are likely to need help from experts on the other side. The case of Danish fisheries has clearly demonstrated that if organizations participating in corporate structures do not have sufficient resources to make independent moves, they will act simply as technical consultants to the public authorities. Hence, the size of the secretariat or the permanent staff of organizations may be an operational indicator of the degree of participation of a particular organization in the political decision-making process.

In the following analysis we shall consequently focus on the size of the

permanent staff and relate this to the different types of interaction - representation in public committees, regular, formal, and other sorts of contacts to public authorities - and to the frequency and scope of these contacts.

Selected relationships between organizational resources and direct organizational participation in government are displayed in Table 2. The data certainly demonstrate that organizational resources operationalized by the size of the permanent staff constitute a fairly good predictor of the frequency, scope, and form of organizational participation in public policy-making. No less than 92 per cent of the very powerful organizations, i.e. organizations with a permanent staff of more than 25, participate daily in the work of the public bureaucracy. This rather exclusive group of organizations is, moreover, linked with more than one ministry. In fact, they cover a broad range of ministries, i.e. they participate in the formulation and implementation of various kinds of public policy. And they possess the capacity and strength needed to participate in a vast array of public committees. These large, strong, and all-comprising organizations probably constitute the core group of'pure insiders'. The 'weak insiders' may be found among organizations with less than six people employed on their staffs. They are, not surprisingly, characterized by a more occasional presence in corporate structure. They are primarily linked with only one or a few ministries; and most conspicuously, they lack the manpower required for participating on a wide scale in the system of public committees. Thus, 85 per cent of these smaller organizations are represented only in a few committees.

3.3. Organizational Structures

It was suggested earlier that the specific character of network relations within a given organizational set might determine the access-structure or the pattern of direct participation in public policy-making. This theme can now be explored further. In the literature on corporatism one often finds propositions stating that corporatist decision-making is inextricably related to the existence of strongly centralized peak-organizations exerting a representative monopoly over the sector of the organizational set concerned (Olsen 1978; Schmitter 1974). This may be the case for some organizations in some countries. Yet the existence and functioning of centralized peak organizations should not be considered as preconditions for incorporation into the corporate structure, but rather as possible co-variants to be discovered through empirical analysis (Johansen and Kristensen 1978b). The point is that the degree of centralization within an

organizational set - often embodied in the form of a strongly dominating
peak or focal organization- influences the access structure, but it does not
in itself constitute a precondition for incorporation.

Two interrelated hypotheses can be tested: (1) For highly centralized organizational sets, direct participation in government is through the peak organization. (2) Decentralized organizational sets tend to produce a fragmented access structure; i.e. participation is spread out across the various organizations which constitute the set. It holds true for both hypotheses in question that the scope of validity is restricted to the Danish case. If the Danish case proves to be deviant in comparison with other countries often classified as corporate, it is nevertheless an interesting one. Table 3 sheds some light on these questions.

The manner in which the data are displayed in Table 3 differs from that of the preceding tables. In order to put some flesh on the crude categories and intervals used hitherto, averages have been calculated. It should, in this context, make it easier to secure a more adequate picture of the strength of the organizations and of the profundity of their interchange with the public administration.

One important aspect of organizational structure is captured by means of a simple three-fold typology of organizations based on their hierarchical location. The first type covers the top of organizational hierarchies, i.e. peak-organizations. For all practical purposes these are also umbrella organizations, i.e. they have other nation-wide organizations as members ² Put differently, this type comprises the focal organizations within organizational sets. The next type includes umbrella organizations that are not peak-organizations, i.e. amalgamations of nation-wide organizations which in one way or another are linked with peak-organizations, but which are placed at a lower level of the hierarchy.

The third type consists of individual organizations. They may or may not be related to existing networks, either as members or through other forms of network-relations. The relative spread of these types of organizations within the functional categories set up along the vertical axis in the table may be interpreted as an expression of the degree of centralization and unity within the categories, when compared to the average figures for resources and public contacts.

If one focuses first on the two central organizational sets related to the labour market, i.e. the LO-system (Danish Confederation of Trade Unions (LO)) and the DA-system (Danish Employers' "Confederation (DA)), an interesting pattern emerges. Both sets are characterized by unity in the sense that only one peak-organization exists and dominates each set. Both

sets are remarkably powerful in terms of in-house staff, income, and members. And there is a certain total symmetry in their participation in public policy-making. By the same token, it should be noticed that these two sets or functional categories are most extensively and intensively involved in corporate decision-making, when compared with other categories. Yet there appear to be significant differences regarding their access structure.

The LO-system is rather decentralized in comparison with the DA-system, and for that matter, in comparison with similar workers' organizations in the other Nordic countries (Elvander 1974). Within the LO-system a certain amount of autonomy remains with other umbrella organizations and individual unions, especially with regard to bargaining. Conversely, the DA-system exhibits a more centralized structure (Buksti and Johansen 1977c). Table 3 reflects these differences in the degree of centralization in terms of relevant resources. With respect to the DA-system, the resources in terms of size of secretariat and money are primarily concentrated in the peak-organization, whereas within the LO-system unions have strong resources at their disposal. Accordingly, and this is the key in this context, the access structure varies significantly from set to set. For the DA-system participation in government is performed almost entirely through the peak-organization (DA). In contrast, the access structure between the LO-system and the public bureaucracy is more fragmented. The peak-organization, LO, does not monopolize public contacts. Although the LO self-evidently plays an important role, public participation to a certain extent is delegated to the unions and to the various umbrella organizations in the network.

The organizational set comprising white-collar unions is extremely decentralized - almost atomized. There exists a multiplicity of partly overlapping and rather weak peak and umbrella organizations, and the set as a whole is marked by a great many organizations (Buksti and Johansen 1977c:395). Accordingly, a fairly fragmented access structure is discernible. The pronounced low level of direct participation in government for the set as a whole, however, is due to the lack of unity and the presence of competing organizations within the set. In such a situation the bureaucracy never knows with which organization to co-operate. Moreover, the advantage of legitimization and consensus-building is minimal when there are competitive organizations, and consequently no one of them can be considered representative.

In spite of these variations, the fact that throughout the entire system of
interest groups peak-organizations actually do play a relatively more

important role in corporate decision-making than do organizations placed at lower levels in the hierarchy of organizations should not be overlooked. These findings about the relationship between organizational structure: and involvement in the corporate structure correspond closely to results from other studies (Johansen and Kristensen 1978b).

We have now shown that organizational participation in corporate decision-making varies across types of organizations, organizational resources, and structures. The extent to which organizational participation is related to specific political issue areas still remains to be examined.

3.4. Direct Group Participation and Political Issue Areas

Not all issues or questions are equally suitable for corporate decisionmaking. The literature has pointed to the fact that predominantly narrow, technical, divisible, and measurable questions fit well into the bargaining process and compromises associated with the corporate structure. Such issues seem to foster solutions in which the participants may be treated as 'sharers' rather than 'winners' or 'losers'. Conversely, ideological or moral questions, or general issues related to basic principles about the political and economic system, seem to belong more to the realm of the electoral-territorial channel (Heisler 1974; Olsen 1978). Indeed, the benefits of efficiency, specialization, and compromise can be maximized more easily. This type of issue, furthermore, makes it easier to minimize the costs of loss of organizational freedom and autonomy, and prevention from claiming pure ideological positions. Moreover, the extent to which public activities and decisions interfere or converge with the field or operation of well-established and recognized interest groups appears to be rather important. If this is the case, the more indispensable will direct organizational participation be seen to be, both from the perspective of the individual interest groups and that of the bureaucracy.

In order to explore these working hypotheses it would be appropriate to construct a typology of public activities based on the above-mentioned dimensions. The difficulties involved in such a task are, however, rather frightening, if not insurmountable (cf. the difficulties discussed by Damgaard 1977). In what follows we adhere to a typology adopted by Grønnegård (1978). It has apparently proved to be useful in similar contexts (Johansen and Kristensen 1978b). The Danish ministries are divided into four functional categories according to their activities:

1. Universal regulation.

This type of public task aims at regulating the behaviour of the citizens,
and the regulations are directed towards the public at large. The Minis-

tries of Justice, Taxes, and Environment could be mentioned as typical
of this class.

2. Specific regulation. Activities of this kind also aim at regulating the behaviour of citizens but, in contrast to the first type, they are directed towards specific sectors of society. The traditional sectoral Ministries are Tråde, Agriculture, Fisheries, Labour.

3. Production of public goods and services. Ministries that provide public goods and services for the citizens are included in this type. The Ministries of Social Affairs, Education, the Interior and Defence constitute the bulk of the service-producing welfare-sector.

4. Coordination of the public administration. A fourth type of public activity is comprised of the work of those ministries that perform cross-sectoral and coordinating tasks between the various ministries within the bureaucracy itself (Foreign Affairs, Finance, Prime Ministers office).

Corporate interest representation may be expected to reach the highest level within the area of specific regulation. For one thing, the questions dealt with here are most often of the narrow, technical, and divisible character so suitable for corporate decision-making. Moreover, these activities substantially penetrate the fields of interest and operation of the big, well-established and recognized organizations related to the labour market and business. If anything, this may be the core-area for the intimate clientele-relations between the organizations of the labour market and the Ministry of Labour, between the organizations of agriculture and the Ministry of Agriculture, between the various trade organizations and the Ministry of Trade, and so on. Public activities aiming at the production of goods and services do not correspondingly interfere with the lives of well-established and recognized interest groups. Consumers of public services are rarely organized as consumers, and if they are, such organizations tend to be weak. One may, consequently, expect a relatively low level of external group participation within this field. On the other hand, large labour market organizations may be expected to direct their activities also towards this area, in part because the core reason for their direct participation in government, i.e. specific regulation, does not exist and function in a political and economic vacuum, but rather is closely related to the various outputs produced by the welfare sector. It may also follow, in part, from a general tendency towards organizational imperialism by the strong peak-organizations in the labour field. Finally,

several white-collar groups representing various categories of public
employees may be expected to be involved in the performance of this kind
of activity.

Table 4 shows the relative number of contact-ministries within the four types of public activities for various functional categories of interest groups. It should be recalled that the unit of measurement in the table is the contact-ministry, not the individual interest group, i.e. one single group may at the same time have contact with different ministries. This should provide a picture of the actual concentration of the public contracts of interest groups across types of issue areas.

As expected, specific regulation constitutes the core-area for external group participation in government. And the organizations of workers and employers, together with those of business, agriculture, etc., are most heavily involved in this kind of public policy-making. Yet the core-ministry varies. Organizations related to the labour market are, not surprisingly,-

primarily involved with the Ministry of Labour, whereas the business
and agricultural organizations are associated overwhelmingly with
the Ministry of Agriculture and the Ministry of Trade.

White-collar groups, by contrast, seem to concentrate their efforts on the production of public services, as noted earlier. In a certain sense this is a form of internal group participation, because the organizations concerned represent the very producers of public goods and services, i.e. public employees. The contacts of white-collar groups with the Ministry of Justice (universal regulation) and with the Ministry of Finance (coordination of public administration) probably have to be looked upon in the same way. In short, to the extent that white-collar groups participate in these areas it is as personnel-groups.

Conversely, the participation of organizations of workers and employers in these fields (Universal regulation and Coordination of the Public Administration), which are otherwise marked by a certain administrative autonomy, is probably of a more 'genuine' external character. The participation of business and agricultural organizations in the field of coordination of the public sector displays similar propterties, being related to the commercial export activities of the Ministry of Foreign Affairs. In a way this is a form of specific regulation - though at the international level. The table also shows that the big organizations of'Labour and Capital' participate in the field of public services as a form of external group participation. It is noteworthy that the Ministry of Education is the most frequent - and most important - contact-ministry for roughly all functional categories of interest groups.

As far as group participation in universal regulation is concerned, the Ministry of the Environment constitutes the most important contactministry. This is because this relatively new ministry has, from the very beginning of its existence, internalized the generally prevailing norm in the bureaucracy about close co-operation with all recognized interest groups affected by the work and tasks of public officials.

4. Conclusions

A dominant tendency in the rapidly growing literature on corporatism seems to proceed from the assumption that there is a special relationship between the structure of interest groups and corporate interest representation. Thus it is often asserted that interest groups participating in public policy-making are organized into a 'limited number of singular, compulsory, non-competitive, hierarcically ordered . . . (organizations)'. Accordingly,

the archetypal 'corporate' interest group is allegedly constituted
by the bureaucratized and strongly centralized peak-organization
exerting a representative monopoly over its functional sector.

The purpose of this article has been to shed some light on these assumptions with the aid of a data base containing information on as many constituent units within the Danish system of interest groups as possible. It is evident that large scale empirical analysis in principle is required if one wants to test hypotheses such as those indicated above.

Direct organizational participation in government in Denmark is certainly not restricted to a small number of interest groups. On the contrary, a very large number of interest groups has contact in one way or another with the bureaucracy. Yet there are, admittedly, great variations in the intensity and scope of the interplay between interest groups and public bureaucracy. These variations can be seen across types of interest groups, organizational resources and structures, and political issue areas.

The empirical findings show that functional interest groups dominate the corporate structure, when compared with so-called promotional groups. But some of the latter do in fact participate, and some do so to a considerable extent. Although the very powerful functional organizations, such as the organizations of workers and employers, play a unique and somewhat symmetrical role in the corporate structure in terms of intensity and scope, they are characterized by different access structures. Most conspicuously, the de-centralized structure of the trade union movement tends to be related to a rather fragmented access structure, whereas the centralized structure of the employers' organization implies that the interchange with the bureaucracy is performed almost entirely through the peak-organization. The existence of strong and centralized peak-organizations cannot therefore adequately be considered as a precondition for corporate decision-making. In that respect unity, non-competitiveness and representativeness within functional categories appear to be more important. The relatively low level of direct group participation by white-collar groups supports this finding.

On the whole it may be said that the Danish case shows that corporate structures and practices go hand in hand with a system of interest groups which is characterized by myriads of interest groups and an over-all, rather de-centralized structure.

To the extent that white-collar groups participate in public policy-making, they do so from within the policy-making subsystem. White-collar groups primarily act as representatives for the various categories of public employees engaged in the production and supply of public goods and

services within the public sector itself. The organizations of blue-collar workers and employers, on the other hand, dominate the field of specific regulation in particular. But their role in the corporate structure is special. They also participate directly in other fields of public activities - and they do so as external participants. Even if their role may be said to be unique in almost all respects with regard to corporate decision-making, in no way do they monopolize the system.

---

NOTES

1. Only organizations on which we have full information are included.

2. Some peak organizations, however, are not umbrella-organizations. The Danish Bankers' Association, The Danish Saving Banks Association, and The National Association of Local Authorities are illustrations.

REFERENCES

Beer, S. H. 1958. 'Group Representation in Britain and the United States', The Annals of the
American Academy of Political and Social Science.

Bentley, A. F. 1908. The Process of Government. A Study of Social Pressures. Chicago:
University of Chicago Press.

Buksti, J. A. 1976. The Adaptation of Organizations to Environmental Changes. The Case of
the Danish Agriculture and Fisheries. Some Tentative Considerations, paper, 2nd EGOScolloquium,
June, 1976.

Buksti, J. A. 1979. Corporate Structures in Danish EC Policy: Patterns of Organizational
Participation and Adaptation. Paper for ECPR-workshop, Brussels, April, 1979.

Buksti, J. A. and Johansen, L. N. 1977a. The Scope and Validity of the Corporatism-Model,
paper for ECPR-workshop, Berlin, April, 1977.

Buksti, J. A. and Johansen, L. N. 1977b. Danske Organisationers hvem-hvad-hvor.
Copenhagen: Politikens Forlag.

Buksti, J. A. and Johansen, L. N. 1977c. 'Organisationssystemet i Danmark', Økonomi og
Politik no. 4, 345-415.

Dahlerup, D., Jarlov, C, Johansen, L. N., and Kristensen, O. P. 1975. Korporatismebegrebet
og studiet af samspillet mellem politiske institutioner', Økonomi og Politik

no. 4, 317-344.

Damgaard, E. 1977. Folketinget under forandring. Copenhagen: Samfundsvidenskabeligt
Forlag.

Eldersveld, S. 1958. 'American Interest Groups: A Survey of Research and some Implications
for Theory and Method', pp. 173-196 in Ehrmann, H. W., ed., Interest Groups on
Four Continents. Pittsburgh: University of Pittsburgh Press.

Elvander, N. 1974. 'Staten och organisationerna på arbetsmarknaden i de nordiska länderna.
En komparativ översikt', pp. 57-131 in Brantgärde, L., Elvander, N., Schmidt, F., and
Victorin, A., Konfliktlösning på arbetsmarknaden. Lund; CWK Gleerup Bokförlag.

Finer, S. E. 1966. Anonymous Empire. London: Pall Mall Press.

Finer, S. E. 1973. 'The Political Power of Organized Labor', Government and Opposition.

Grant, W. 1977. Insider Groups and Outsider Groups. An Exploration of some Factors
Affecting Interest Group Strategies in the United Kingdom. Paper for ECPR-workshop,
Berlin, April, 1977.

Grønnegård Christensen, J. 1978. Normerne for samspillet mellem den danske centraladministration
og interesseorganisationerne i centraladministrativt perspektiv. Paper, Fifth
Nordic Political Science Conference, Bergen, August, 1978.

Heisler, M. 0., with Kvavik, R. B. 1974. 'Patterns of European Politics: The 'European
Polity' Model', pp. 27-89 in Heisler, M. 0., ed., Politics in Europe: Structures and
Processes in Some Postindustrial Democracies. New York: David McKay.

Jarlov, C, Johansen, L. N., and Kristensen, O. P. 1976. The Danish Committee System: A
Preliminary Report from a Study of Danish Committees 1946-1975. Paper, IPSA Congress,
Edinburgh, 1976.

Jessop, B. 1978. Corporatism, Fascism and Social Democracy. Paper for ECPR-workshop,
Grenoble, April, 1978.

Johansen, L. N. and Kristensen, O. P. 1978a. Dataarkiv indeholdende oplysninger orn
offentlige udvalg 1945-75. Institute of Political Science, University of Aarhus, Denmark

Johansen, L. N. and Kristensen, O. P. 1978b. Corporatist Traits in Denmark 1946-76. Paper, ISA Conference, Uppsala, August, 1978. To appear in Lehmbruch, G. and Schmitter, P. C, eds., Consequences of Corporatist Policy Making. SAGE Modern Politics Series vol. 5, London and Beverly Hills: SAGE Publications (forthcoming).

Latham, E. 1952. The Group Basis of Politics. Ithaca: Cornell University Press.

Lehmbruch, G. 1977. 'Liberal Corporatism and Party Government', Comparative Political
Studies vol. 10, 91-136.

Meijer, H. 1969. 'Bureaucracy and Policy Formation in Sweden', Scandinavian Political
Studies vol. 4, 103-116.

Moren, J., ed., 1974. Den kollegiale forvaltning. Oslo: Universitetsforlaget.

Olsen, J. P. 1978. Integrated Organizational Participation in Government. Unpublished,
Bergen. To appear in Nyström, P. C. and Starbuck, W. H., eds., Handbook of Organizational
Design.

Panitch, L. 1977. 'The Development of Corporatism in Liberal Democracies', Comparative
Political Studies vol. 10, 61-90.

Peterson, E. A. 1977. Interest Group Intercorporation in Sweden. Paper for the 1977 Annual
Meeting of the APSA, Washington.

Schmitter, P. C. 1974. 'Still the Century of Corporatism?', pp. 85-131 in Pike, F. B. and
Strich, T., eds., The New Corporatism. London: University of Notre Dame Press.

Schmitter, P. C. 1977. 'Model of Interest Intermediation and Models of Societal Change in
Western Europe', Comparative Political Studies vol. 10, 7-38.

Terreberry, S. 1978. 'The Evolution of Organizational environments', Administrative Science
Quarterly, 590-613.

Truman, D. 1951. The Governmental Process. Political Interests and Public Opinion. New
York: Alfred A. Knopf.

Wilson, J. Q. 1973. Political Organizations. New York: Basic Books Inc.