Resumé

This article investigates the pattern of corporate political representation in Danish
law-making. The most important finding is a large intra-system variation in
scope, level, and structure of ouiside involvement. There is no evidence of an
emerging 'total' system of corporate political representation. The dominant
pattern is still one of a diversified use of several channels of access by a large
variety of groups, organizations, and institutions. While the importance of organized
interests is well documented, it is questionable to what extent this indicates
a pattern of societal corporatism in the political system. Interest organizations
play an important role lin pluralist societies as well, and Philippe
Schmitter's main point about representational monopoly does not gain support.
Generally, the analysis tends to support a model of pluralist corporate representation.

1. Purpose and Design

1.1. Introduction

Recent years have witnessed a growing concern for the relationship between government and organized interests. Several authors argue that corporate political representation has assumed an increasing importance in Western political systems. Interest organizations' participation in formulation and implementation of decisions has become a major characteristic of political systems. A large number of boards and commissions with interest representation have been established, and patterns of frequent consultation between organized interests and governments have developed. Similarly, the importance of parliament is allegedly reduced while the role of government and administration has expanded.

These assumed characteristics of the decision-making structure have changed not only the focus in studies of interest representation, but also the concepts and paradigms used in analysis of the relations between state and society.

---

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

The analytical frameworks and theories developed by scholars in the field vary considerably, however. Some authors stress the Unking structure between interest organizations and government offering a 'new' European Polity Model (Heisler, 1974). Some are more concerned with the structural properties of the interest organization system, as, for example, Philippe Schmitter, in his analysis of 'state and societal corporatism' (Schmitter, 1974). Others emphasize the increasing importance of collective actors in the 'corporative bargaining channel' as alternative to individual representation through the 'numerical' channel. The value of numbers - votes - as a political resource has declined while the value of organizational resources has increased. The concept of corporate pluralism has been suggested to describe this type of representation (Rokkan, 1966).

Most empirical studies of interest representation have focused on a few areas of public policy (Elvander, 1972), on one particular phase in the decision-making process, e.g. government boards and commissions (Moren, ed., 1974) or on one group of outside actors, most notably interest organizations (Elvander, 1969; Kvavik, 1976). But the total structure of external involvement and the relative importance of numerical and corporate channels of interest representation have so far rarely been dealt with in empirical analyses.

Even less empirical work has been done on the participation of external actors other than interest organizations. One of the reasons for this state of affairs is that in most studies the focus has usually been on organized interest intermediation and not on the functioning of the political system as a whole. Most theories and hypotheses of the functional representation have been developed in relation to the involvement of organized interests. The general assumption is that organizations dominate external involvement in the policy making and implementation processes. But few have tried empirically to establish the relative importance of organized interest representation at the various stages of the decision-making process.

Another aspect is the structure of the participation by interest organizations: To which extent is there monopoly or exclusiveness in the representational system? How is the organizational network structured in various sectors of the society? Philippe Schmitter deals mainly with such aspects of the decision-making process when he defines corporatism as:

'a system of interest representation in which the constituent units are organized
into a limited number of singular, compulsory, non-competitive, hierarchically
ordered and functionally differentiated categories, recognized or licensed (if not

created) by the state and granted a deliberate representational monopoly within their respective categories in exchange for observing certain controls on their selection of leaders and articulation of demands and supports'. (Schmitter, 1974:94-95).

He implies that one will find a. limited number of organizations involved in
the different areas of legislation at the administrative preparation and
implementation stages.

The aim of this study is to investigate the patterns of external involvement in the Danish law-making process. 'External' involvement refers to participation by organizations, institutions, and individuals in contradistinction to 'internal' parliamentary and governmental participation. The analysis covers all stages in the preparation, deliberation, and implementation of Government bills passed in 1972-1974.

We attempt to answer three questions in this paper: 1. To what extent do external actors participate at the different stages in the law-making process? 2. What is the relative importance of interest organizations in the various phases? 3. To what extent is representation restricted to a limited number of actors?

The first question leads us to investigate two main aspects: First, the scope of outside involvement in the law-making process, i.e. the number of bills with participation (at the administrative and parliamentary stages) by interest organizations and other external actors. Second, the level of participation in relation to each bill, as measured by the number of'acts' or 'units' of participation. The first dimension concerns the range of issues that organizations, institutions, and individuals are involved in. The second focuses on the frequency of participation with regard to different bills.

The second question covers both a) the proportion of the bills with interest organization participation, and b) the number of'acts of participation' by organizations as a percentage of the total number of acts of participation by non-governmental actors. By employing these measures we attempt to assess the relative importance of organized interests in the various phases of the law-making process.

Finally, the degree of concentration referred to in the third question will
be measured by the number of acts of participation per actor at the
different stages.

We expect that the answers to all three questions will show variations
across areas of public policy The content of specific decisions presumably
influences the level and scope of outside involvement, the relative

importance of organized interests, and the degree of concentration in the
representational structure.

1.2. Framework and Data

Ideally we would like a total mapping of participation patterns in Danish policy-making. But this is of course not feasible. The problems of identifying 'all' relevant decisions and of obtaining complete information on the behavior of actors are overwhelming, indeed insurmountable. Therefore we restricted the analysis to Government bills passed by Parliament. This population of decisions is clearly defined, it covers almost all areas of public policy, and affects most sectors of society.

Furthermore, we had to restrict the analysis in time. We selected the two parliamentary sessions of 1972/73 and 1973/74 which gives us a reasonably large and varied population of laws for investigation of intrasystem variation in participation patterns. However, we are not able to analyse long termchanges in the involvement of external actors. Thus, we cannot test hypotheses about the causes of the present state of affairs or trace the origins of the observed patterns of interaction.

Finally, it should be emphasized that we had to usepublic documents as data source, which means that only officially reported acts of participation by organizations and individuals can be studied. Obviously, there are numerous informal contacts and influence relations which we cannot unravel because they are not reported in such documents. Formal participation in commissions and administrative boards as well as contacts with parliamentary committees are well documented. The major problem is that there is no easy way of knowing whether the ministries apply similar guidelines concerning the extent to which they report consultations with organizations and institutions.

The data used in this paper are stored in an archive of bills proposed in
the Danish Parliament in the period of 1972-1974 (for detailed information
on design and coding, see Damgaard & Eliassen, 1975, 1976). The total

number of units in the archive is shown in Table 1. Presently we are only using the 298 Government bills proposed and passed by Parliament in the two completed parliamentary sessions (1973/74 I was a short session leading up to the general election of December, 1973).

For each unit (bill) available information has been coded concerning:

1. Identification, classification, and content

2. Stages the bill passed and how it was håndled at each stage

3. Who participated - institutions, organizations, individuals - at each
stage.

For analytical purposes the legislative process can be divided into 9 steps or stages which a bill can pass during preparation, deliberation, and implementation. The nine steps used to structure the data on each bill are the following:

1. The commission stage

2. The administrative preparation stage

3. Introduction in parliament

4. First reading

5. Parliamentary committee stage

6. Second reading

7. Second parliamentary committee stage

8. Third and final reading, including final division

9. Administrative implementation

The actual number of stages a bill passes depends, of course, upon the type
of bill, its origin, and how it is handled by Parliament. For example, most
private members' bills are killed at the first committee stage.

External participation by organizations, institutions, and individuals has been coded for five of the nine steps: commission stage, administrative preparation stage, first and second parliamentary committee stages, and administrative implementation stage. The data on involvement are organized for each 'act of participation' to include information on a) the stage in the process, b) the institution/organization participating, and c) the individual persons involved.

The main file in the archive is then a file with bill as the unit including all
information regarding the bill (identification, content, handling of bill) and
each 'act of participation' described by the three variables mentioned.

By using this main file we have constructed two additional files with 'act

of participation' as unit: one for organizations/institutions involved and one for the persons participating. In these two archives a new unit is created each time an organization and/or a person participate at any stage in the process. These files allow investigation of the participation structure of each actor or group of actors. From the master file of bills additional information on the content of bills is transferred to the two new files. The relations among the three files are shown in Figure 1.

2. The Dimensions of Law-Making: Actors, Stages, Content

2.1. The Actors

The questions formulated in the previous section imply a three dimensional
matrix for the analysis: actors, stages in law-making, and content of

bill. Furthermore, we have two main 'levels' of analysis: the macro or system level and the organizational/institutional level. At the system level our unit of analysis is bill or group of bills. But investigation of the intensity of participation requires a shift to the organizational level. The unit of analysis is then act of participation by organizations and institutions.

In addition, there is a third 'level' of analysis, viz. the individual level. Individuals can participate either as private persons or as representatives for organizations and institutions. The unit is then act of participation by individuals.

The distinction between organizational and individual actors is necessary in order to achieve an accurate picture of external participation. Individuals are involved in various capacities and in some cases their organizational or institutional affiliation will not be indicated in the official documents. Thus, participation of individuals without such affiliation has to be included in order to get a complete account of the scope and level of external participation.

For most of the analysis reported in this paper we will use a fairly simple categorization of external collective actors: a) interest organizations; b) other non-governmental actors, including municpalities, corporations, parties, individuals, and ad hoc groupings of individuals, c) governmental institutions, including agencies, boards, and commissions. Most of the actors can easily be categorized according to this scheme. The residual category contains less than one per cent of all external actors.

2.2. Stages in Law-Making

The second dimension in our matrix is the stages a bill passes from initiation to final implementation. As previously indicated, four stages are of particular importance: the commission stage, the administrative consultation stage, the parliamentary committee stage, and the administrative implementation stage.

These four stages and the corresponding channels for external involvement have been analyzed by several authors. First, commissions or similar bodies in which interest groups and various types of experts are represented have become increasingly important at a very early stage of the drafting of legislation (Moren, ed., 1974; Meijer, 1969; Offentlige udvalg, 1970; Jarlov, Johansen, & Kristensen, 1976). Second, formal or informal systems of mutual consultation between government and administration on the one hand and interest organizations and professional experts on the other have been increasingly used in the administrative preparation of bills

(Elvander, 1969; Kvavik, 1976; Ingebrigtsen, 1968). Third, a declining importance is attached to Parliament by interest organizations as compared to consultations with Government and bureaucracy (e.g. Melander, 1974). Elvander's interviews with Swedish organization leaders confirm this pattern. His results show that organizations generally attach very little importance to contacts with Parliament and parliamentary committees and a correspondingly high importance to administrative hearings and, in particular, membership of commissions in almost all areas of public policy (Elvander, 1969:253, 255). The same preference pattern is found in Norway for all areas except agriculture and fisheries (Moren, ed., 1974:30). Fourth, increasing use of quasi-governmental agencies for the execution of legislation and direct participation of interest organizations in the preparation and implementation of laws have been reported (Hauge et al., 1975; Heisler, 1974:62).

As noted above, we are not able to analyse changes in the use of different channels, but we will try to describe the end-product of these developments in the case of Denmark with the ultimate aim of assessing the contemporary importance of the four channels.

We assume that variations in the external involvement across the channels of participation are mainly due to the varying accessibility of the channels. Some are simply more closed than others. The creation of legislative commissions and of extra-departmental bodies for the implementation of laws is subject to a decision by Parliament or the minister in charge. The membership of these bodies is determined by the same authorities, although traditions have evolved as to who should be included in commissions and participate in implementing laws. The composition of such bodies is partly a result of a long term adjustment between the state and outside groups. Administrative consultation is also a channel created by the government. Decisions on who should be consulted are formally made by the responsible minister. In contradistinction, access to parliamentary committees is open to everybody.

Earlier studies indicate that interest organizations prefer, if possible, to participate at the commission and administrative consultation stages rather than at the parliamentary committee stage. This fits the assertion of several authors that Denmark, and Scandinavian countries in general, can be characterized as corporate societies. A shift should have taken place from lobbying in Parliament to participation at the initial preparation and final implementation stages.

But it is not unlikely that the access structure and the relatively decentralized
and fragmented organizational system in Denmark (Buksti &

Johansen, 1978) actually limit tendencies towards corporate patterns of
political representation.

2.3. Content of Bills

The third dimension in our data matrix is the content of decisions made. Content is important because it is expected to influence political processes as Theodore I. Lowi (1964), Lewis A. Froman (1968), and others have emphasized. Unfortunately, attempts at applying content classification schemes have been few and generally not very successful because of severe problems of operationalization and measurement (see, for example, Jones, 1977; Volgler, 1974; Ripley & Franklin, 1976; Damgaard, 1977). Nonetheless, we assume that external participation is dependent upon the concrete subject matter.

This approach implies that laws differ in important respects. We therefore ought to ask a preliminary question: Are laws at all comparable units with respect to external participation? If not, which laws are 'normal' or 'typical' and which are 'deviant' so that external participation cannot be expected to occur, or perhaps only be expected to take place in very specialized ways? Our answer to this basic question was to review the universe of bills passed and to define several categories which should be excluded: naturalization laws, laws applying only to Greenland or to the Faroe Islands, laws postponing amendments otherwise planned, laws adjusting existing legislation as a consequence of other bills passed or of European Community policies.

Table 2 shows the number of Government bills passed in the various

categories (which are almost identical to those used in Damgaard, 1977:
ch. 9). In the following analysis our data base will thus be 191 'normal
laws'.

Aside from the exclusion of'deviant' legislation the content of bills will be
described by two variables in this paper: a) financial consequences for
government, and b) policy area.

An important aspect of the relationships between state and society is that the government controls both the extraction of financial resources from society and the distribution of resources to various sectors of society. A major objective of interest groups is to influence this reallocation process, to the benefit of their members and more generally to the benefit of 'their' sector of society. Our general hypothesis is that the scope and level of external involvement in the law-making process is higher in the case of bills containing proposals for increased government spending than with regard to bills increasing the extraction of financial resources. We expect that spending bills will more often be subject to hearings and preparation by commissions than will revenue bills which, on the other hand, might show a high degree of external involvement at the parliamentary committee stage. Furthermore, we expect that the actors involved concerning spending bills are predominantly interest organizations, while a large number of individuals, business firms etc. participate concerning revenue bills, particularly to voice criticism at the parliamentary committee stage.

To analyze these relationships all normal bills have been categorized according to their financial consequences: 1) bills increasing government revenue/decreasing government expenditure, 2) bills increasing government spending/decreasing revenue, 3) bills with zero effect on government spending/re venue. An additional fourth category is used for bills on which information is inadequate. Ideally, one would like a less crude measure of financial consequences, but this reflects the (generally poor) quality of the official documents submitted to Parliament by the Government.

The second content variable is policy area of bill. Most authors agree that there are significant differences with respect to interest organization involvement across policy areas. There is also some agreement on the main causes of these variations. The aspects most often mentioned are a) the presence of clear standards for definition and measurement of the outcome of decisions (Olsen, 1976:72), b) the degree of organizational 'coverage' of the different areas, c) the possibility of defining a limited number of organizations, perhaps only one, 'responsible' for the area (Schmitter, 1974:94-95), and d) the traditions and rules applied within the

area in question. According to these four criteria there are large variations across policy areas, no matter how they are defined. Most authors agree, however, that at the one extreme one finds in most countries some kind of an industrial/economic/clientele oriented area with a) clear standards for defining the issues at stake., b) a high organizational coverage, c) a few or only one 'top' organization, arid d) old traditions and well established rules for at least an active consultation in the policy-making process. In several countries we find establishment of the first formal state-interest organization-expert bodies for consultation on policy issues in this area.

We would therefore hypothesize a much higher extent of corporate participation patterns at the commission and administrative preparation stages in this area than in others. The problem is to define the area more precisely. The general notion of economic policy is too broad, including as it does strictly financial issues in which organizational participation is expected to be almost reversed. On the other hand 'socio-economic' or 'industrial' policy differs from areas less directly oriented towards economic distribution and redistribution in society, as, for example, environment and education. In these areas the interest organization structure is less developed and more fragmented and there are less clear standards for defining possible solutions to the problems dealt with.

One procedure which allows us both to distinguish between financial and industrial socio-economic policy, and to create two relatively homogeneous residual areas, is to introduce the common distinction between traditional and modern areas of state activity. The traditional areas cover mainly finance, foreign affairs, defence, justice, and, in particular in protestant countries, the cluster of education, church, and culture. The modern areas of public activity are mainly directed towards intervention in economic and industrial matters, the establishment of a social welfare system, and state involvement in environmental protection, housing etc.

The two dimensions give us a simple fourfold classification of policy areas. The remaining problem is to group the bills according to the scheme. As a tentative solution, we shall use the minister presenting the bill to Parliament as classification criterion and then group the ministries - admittedly very roughly - according to a) the timing of establishment of state activity in the area, and b) the degree to which financial or economic policy is involved. The result is four main areas of public policy and the corresponding ministries (M):

So far we have presented two variables relating to the content of bills. The final question is how and to what extent they are interrelated. Table 3 shows the relationships between financial consequences of the bills and the four areas of public policy.

The table points first of all to the differences between financial and industrial policy. Industrial policy entails predominantly either bills increasing government expenditures or bills with no financial effect on the government budget. This supports the argument about industrial policy as an area of state regulation and economic involvement.

Secondly, we find a striking difference between the two residual categories. There is a high proportion of'no financial effect' bills in the traditional areas of state activity and a high proportion of bills with net government expenditure in the category of 'other modern' areas.

3. The Overall Pattern of Participation

3.1. The Scope of External Involvement

In this section we investigate; the participation of interest organizations and other external actors at the four stages in the legislative process. The data show that 160 normal bills (out of 191), or 84 per cent, had been subject to involvement by outside actors at one or more of the three preparatory stages. The result indicates an extensive range of participation by actors outside the framework of the legislature. But if corporate theory implies that the parliamentary channel is used less than other channels, then the theory is clearly not confirmed.

Table 4 shows that two thirds of the laws have caused contacts to parliamentary committees compared to 60 per cent that led to outside participation at the two administrative preparation stages. The traditional pattern of interest articulation through the parliamentary channel is still as commonly used in Denmark as is the 'corporate' representation pattern.

This finding does not exclude a possible shift over time in the use and importance of the different channels of access to the legislative system. The ultimate questions are: Who participates at the different stages, and what is the effect on the decisions made?

One possibility is that interest organizations dominate at the two administrative stages while the parliamentary channel is used by actors without access in the commission or consultation stages. However, a look at the proportion of bills for which interest organizations have participated rules this possibility out. Table 5 shows that interest organizations participated at the administrative preparation stages concerning 41 per cent of the bills but with regard to 59 per cent at the parliamentary committee stage. Thus the pattern of a more extensive scope of outside involvement at the parliamentary committee stage holds for organized interests as well.

If we look at the two administrative preparation stages separately, the consultation stage shows a much lower degree of participation by outside actors (46 per cent) than the parliamentary committee stage. The lowest degree is found with regard to commissions which were established in 31 per cent of the cases for the preparation of bills. The differences between the three preparatory stages are even more pronounced if we look at the proportion of bills in which interest organizations were involved. Commissions turn out to be the most restricted channel of influence with regard to both the proportion of bills prepared in this way and the extent to which extra-governmental actors are allowed to participate. More than one-third of the commissions had members from various governmental agencies only. Representatives from external groups and institutions other than interest organizations were included in commissions for only 7 per cent of the bills.

Turning to the administrative implementation side, we find a relatively wide scope of external involvement: In 50 per cent of the cases special administrative bodies (councils, boards, committees) had been established in order to implement the provisions adopted, or to assist in the executive process. Interest organizations were represented insuch administrative bodies with regard to 92 per cent of these bills.

This rather wide scope is not a perfect indicator of the importance of external involvement at the implementation stage, however. It is also important to know who the external members of the administrative bodies are, and the frequency of their participation.

3.2. Types of Participants

So far we have investigated the scope of external involvement. In order to
study the level of participation at the different stages we have to use the
single act of participation as unit of analysis.

The difficulty with these data is that in the commission and implementation phases one finds organizations with several members in collective bodies which means that several acts of participation will be counted for such organizations in each case. To avoid this problem we have counted only one act of participation for each actor for each bill at the two stages.

Table 6 shows the total number of acts of participation by six groups of actors in the four phases of the legislative process. The important role of interest organizations in all phases of the process is more visible now compared to the scope analysis of external involvement. Organized interests account for about 58 per cent of all acts of participation by nongovernmental actors.

The second largest group of collective actors is governmental institutions not directly responsible for the legislation in question: Other ministries, departments, and agencies that are consulted or are represented in boards and commissions.

A substantial number of individuals participate at the commission and, in particular, at the administrative implementation stage. In some cases their status is not quite clear, but preliminary inspections suggest that many function as technical experts and legal advisors. The few individuals involved at the parliamentary stage appear predominantly to be ordinary citizens.

Business firms, local governments and ad hoc groups of citizens are involved to a significant extent only in the parliamentary committee phase. At the two other preparatory stages the few corporations involved are mainly national cooperative business firms and similar organizations which in some respects could be viewed as interest organizations. The municipal institutions participating are mainly the two communes outside the national organization of local governments, Copenhagen and Frederiksberg, and municipal corporations. Only at the parliamentary stage do we mainly find private firms and ad hoc groups of individuals.

These findings could indicate that organized interests are the main actors in the phases of limited access, thus supporting the idea of corporate political representation. Parliament represents the only major channel for groups outside the organizational framework, or for groups in opposition to the policy of their organization, to present their views.

On the other hand, interest organizations are almost as important in the parliamentary committee phase as in the other preparatory phases. And as indicated in Table 7, the number of acts of participation by interest organizations as a percentage: of the total number of acts of participation by non-governmental actors is significantly higher only in the administrative consultation phase.

The general conclusion is this: To the extent that there is collective outside involvement in the various phases, interest organizations are the main actors. The question is whether this picture will be modified when the content of bills is taken into account.

Our final question is to what extent access to the law-making process is restricted to a limited number of organizations, and whether the degree of concentration varies from one stage to another. Table 8 shows - with exception of the administrative implementation phase - that the number of acts of participation per actor is about 2. There are no signs of a high degree of concentration and the differences between the three preparatory stages are moderate. The only major deviance exists at the administrative consultation stage at which the concentration is higher.

In sum, the overall pattern of external participation in Danish lawmaking is characterized by 1) a high degree of outside involvement, 2) the importance of organized interests, 3) few signs of concentration in organizational participation, and 4) a rather equal use of administrative and parliamentary channels at the preparation stages. Interest representation in the Danish political system seems to be more differentiated and fragmented than most students of corporatism or the new European polity model have assumed.

The question then is whether this pattern hides substantial variations
across public policy areas and types of decisions. We will turn to this
question in the next section.

4. Financial Consequences and External Involvement

First we shall investigate the effect of financial consequences for the scope of external involvement. Table 9 shows the proportion of bills with one or more instances of external participation in general, while Table 10 supplies similar information with respect to interest organizations in particular.

The hypothesized broadest scope of participation with regard to laws increasing net spending receives some support, particularly in the commission and implementation stages. The support is strongest in the case of organizational participation (Table 10) for which the argument was mainly made.

Contrariwise there is a very low degree of outside involvement with laws increasing government revenue at the first two stages. But at the parliamentary committee stage the participation is above average. The process appears to be like this: Prior to introduction of bills in Parliament, preparations are mainly an internal matter for the Government and administration (and perhaps political parties outside Government) because plans to increase taxes or to introduce new duties etc. have to be kept secret as long as possible. Therefore, extensive preparations in commissions with participation by outside groups are precluded. It is also likely

that interest organizations wamt to retain maximum freedom to intervene at the parliamentary stage and not to be coop ted into an agreement at an earlier stage. In the parliamentary committee phase participation is open to everybody that wants to express disagreement or to give advice in public.

If we look not only at the scope of external participation, but also at the frequency of contacts in relation to each bill, the particular character of the net revenue bills stands out even more markedly: At the administrative consultation stage there are only a few bills with 1-5 acts of participation, and zero with more than five:, but at the parliamentary committee stage there are 35 per cent with 1-5, and 38 per cent with more than 5 contacts. As a contrast, the bills increasing Government spending show a limited number of contacts at both of the two stages.

At the implementation sid«::, there is only one major variation from the overall picture. Bills with no financial effect are rarely implemented by special administrative bodies. This could indicate that bills without financial implications can more easily be administered without external involvement, or that external actors are not interested in participating, or that the bills typically belong to the non-economic traditional policy area, cf. Table 3.

The hypothesis of a relatively high degree of external participation with regard to net spending bills gets further support if we look at the level of involvement. If participation occurs, there is a high frequency of outside representation with regard! to these bills compared to all other types of laws. This is true in particular if only involvement of non-governmental actors is considered.

In Table 11 we have shown the proportion of acts of participation by interest organizations as a percentage of the total number of acts of participation by non-governmental actors. Again the main conclusions from the investigation of the; scope of involvement are confirmed. The level of interest organization participation is higher in the case of net expenditure bills than with regard to laws increasing government revenue. On the other hand, the category of no financial effects also shows a high level of participation by interest organizations. This implies that interest organizations are the dominating external actors at the implementation stage although the scope of the involvement is limited (Table 9).

The expected low level of interest organization participation in the parliamentary committee phase for net expenditure bills is not confirmed. On the contrary, interest organizations play a larger role here than at two of the administrative stages. Only in the administrative consultation phase

is there a higher level of organizational participation. In the case of net revenue bills, the findings on the scope of the participation are further confirmed. The level of organizational involvement is somewhat lower in the parliamentary committee phase than at the administrative consultation stage.

Furthermore, the measure of concentration in external participation is at the overall low level in the case of net revenue category in the parliamentary committee phase. Table 12 shows that there is approximately one act of participation per actor in that case. On the other hand, the expectation of high concentration in the administrative preparation phases for net expenditure bills is only met in the commission phase. At the administrative consultation stage the number of acts of participation per actor is not higher than in the parliamentary committee phase.

The findings underscore the open access at the parliamentary stage and the low level of involvement at the two other preparatory stages for net revenue bills. The level of organizational involvement is low concerning these bills at the parliamentary stage, which has the broadest pattern of participation by local governments, business firms, and individuals.

The expected corporate pattern of participation for net expenditure bills is only found to some degree. The level of organizational involvement is high, but the administrative channels are not used more frequently than the parliamentary, and the level of concentration in the representation structure is not significantly higher at the administrative preparation stages. Thus, introduction of the financial dimension gives some additional knowledge of intra-system variation in representation patterns, but the main result is still the highly fragmented and diversified character of the Danish political system and the major importance of outside representation at the parliamentary level.

5. Policy Area and External Involvement

Concerning policy areas the results are only partly in accordance with our expectations. As shown in Tables 13 and 14, the modern industrial area is not a segment of public policy with the most extensive scope of external involvement at the commission and administrative consultation stages. We do not find the high emphasis on the stages of low access which we should expect according to the corporate society hypothesis. But a fuller account of the participation pattern requires analyses of participants and levels of participation at the different stages.

There is an important difference between the two residual categories of public policy. The scope of external participation in the implementation phase is very low in the traditional fields of government activity. Implementation is mainly a matter for either the departments or their subordinate directorates. In the modern areas there is often external involvement in the implementation process.

The major deviation from the overall pattern of participation is found in the area of financial policy. In this area there is a low degree of participation at the first two preparatory stages, and one slightly above average at the parliamentary committee stage. Furthermore, if one looks at the number of contacts concerning each bill, one finds none above 6 in the administrative consultation phase but numerous in the parliamentary committee phase. In this area of public policy, a competitive pluralism model still seems to describe the pattern of outside influence on the decision-making process.

Table 15 shows the number of acts of participation by interest organizations in per cent of the total number of participation units accounted for by non-governmental actors. Using this dimension of involvement, we receive some support for the hypothesized high level of external involvement in the modern industrial area. This type of public policy has the highest total number of acts of participation by non-governmental actors.

Another important finding is the role played by organized interests in the modern industrial area which is larger than in other areas of legislation. At all stages, bills concerning labour market, commerce, industry, and agriculture have the highest level of interest organization involvement. Totally, organized interests account for 90 per cent of all acts of participation. However, we do not find a substantially higher level of organizational representation at the two administrative preparation stages than in the parliamentary committee phase. Thus, lobbying in Parliament turns out to be of importance in this area of legislation as well.

The traditional financial area has the lowest level of external involvement in the administrative consultation and commission phases, but a high degree of outside participation at the parliamentary committee stage. This finding gives further support for the importance of the parliamentary stage. The role of interest organizations is also more restricted in the parliamentary phase with regard to the financial area.

If we look at the participation by other non-governmental groups, one interesting finding is that corporations are main actors in the financial area. They are heavily engaged in all cases involving new taxes or duties at the parliamentary committee stage, but only marginally at the administrative preparation stage. This indicates that to the (limited) extent consultations occur prior to presentation of the bill in parliament, the main actors are relevant interest organizations, but at the parliamentary committee stage individual corporations have the opportunity to present their particular interests to the decision-makers.

Municipalities are involved at the administrative preparation and parliamentary committee stages mainly in the case of the other traditional and other modern areas. But they are not involved with industrial policy, and their interest organizations arc; only marginally involved in this area. This could indicate that local governments consider industrial policy to be the matter for the functional interest organizations and not for the municipalities and their organizations.

So far the general expectation that legislation in the modern industrial area will show corporate patterns of external involvement has received only limited support. The variations in the degree of concentration in the representation pattern is also contradictory to the ideas of Schmitter and others. There are no systematic variations in the number of acts of participation per actor for the four areas of public policy.

Table 16 shows that only in the parliamentary committee phase do we find
the highest degree of concentration in the modern industrial area. At all

other stages, this type of legislation shows a relatively low degree of concentration. By far the highest degree of concentration is found in the implementation phase of the traditional financial area. Actually, 13 different organizations are represented 150 times in governmental boards concerning 24 bilis. However, to a large extent the same agencies are involved for several bilis dealing with taxes and duties. Thus, the excessive degree of concentration is mainly due to a relatively large number of separate tax bilis.

The overall conclusion is still, contrary to what corporate theory would
lead us to expect, that there is no evidence for a relatively high degree of
concentration in the administrative phases of the modern industrial area.

One interpretation of this conclusion could be that the group of industrial bills is defined too broadly. The question is, for example, to what extent labour market legislation would show restricted access to the commission and consultation stages. Our data indicate that it does not. In labour market legislation there are also a large number of actors involved in administrative preparation. The organizational system is highly centralized in the private sector, but often commissions and hearing procedures include not only the two top organizations of the private industrial sector, but also representatives for public employees (with several different organizations), white collar organizations, specialized employers' organizations

in banking, agriculture etc. The result is a more fragmented representation
pattern than scholars advocating corporatism models assume.

6. Some Concluding Remarks

Although this analysis is preliminary, some general patterns of external participation are revealed. The most important result is the large intrasystem variation in scope, level, and structure of outside involvement in the legislative process. There is no evidence of an emerging 'total' system of corporate political representation at the macro level. The dominant pattern is still one of a diversified use of several channels of access by a large variety of groups, organizations, and institutions. The parliamentary committee channel is used as frequently as others by outside actors.

The investigation of the variations in the involvement of external actors yields at least three major findings. First, the large differences among the four stages in the legislative process with regard to scope and level of external participation, type of actor, and exclusiveness of participation. By far the highest degree of outside participation is found at the parliamentary committee stage where access is open to everybody. The most restricted channel is the commissions where external involvement is limited. Few bills have been subject to this type of preparation, and participation in the commissions is restricted to mainly government institutions, certain types of interest organizations, and experts. The degree of concentration in organized interest representation appears to be rather low, but it is difficult to interpret the degree of representational monopoly without a detailed examination of the organizations involved.

Generally, the commission, the administrative consultation, and the implementation stages are domains of the bureaucracy, experts, and interest organizations. The only major channel of possible influence for other actors is the parliamentary committee. This result underscores the importance of organized interests in Denmark, but it is questionable to what extent this indicates a pattern of societal corporatism in the political system. Interest organizations; play an important role in pluralist societies as well. The decisive question is the structure of the relationship between interest organizations and the state.

At the parliamentary stage the role of the bureaucracy is reduced in terms of outside contacts to the parliamentary committee, but there are frequent and important contacts through the minister responsible for the legislation. He is the major channel of bureaucratic influence in this part of the law-making process. One interesting aspect is the low profile displayed

by political parties in the commissions and administrative implementation phases. Other studies indicate a gradual decline in the representation of political parties in various governmental boards and agencies. However, it is sometimes difficult to determine to what extent representatives from interest organizations indirectly act as participants for political parties. One example is the close relationship between the LO and the Social Democratic party.

The second interesting result is the major differences between bills in the traditional financial area and bills increasing government revenue on the one hand, and all other bills on the other. Decisions in the former areas appear mainly to be made within the departments responsible, and the major part of outside involvement takes place at the parliamentary committee stage. All other policy areas and types of financial consequences show a much higher degree of outside participation at tne first two preparatory stages.

Third, the expected high degree of societal corporatism in the modern industrial area has not been found. There is only a weak tendency toward a higher degree of external involvement at the commission and administrative consultation stages in that area. There is a slight tendency toward a higher degree of organized interest representation in the modern industrial area (cf. Table 15), but Schmitter's main point about representational monopoly does not gain support. On the other hand, the general importance of pluralist corporate representation is well documented.

The findings call for a closer look at the organizations participating at the various stages and for alternative conceptualizations of policy content. It is not unlikely that further analyses will reveal more centralized and monopolized structures of representation in specific segments of public policy-making and not in others.

Furthermore, we have to investigate the relationships across the legislative process: To what extent is participation at one stage determined by participation at previous stages? How does the degree and structure of outside involvement affect the handling of bills in Parliament (including the final vote after third reading)? Answers to such questions will also give further evidence on the importance of corporate pluralism in the Danish political system.

REFERENCES:

Buksti, J. A. and Johansen, L. N. 1978: 'Organisationssystemet i Danmark', Økonomi og
Politik 51, 345^15.

Damgaard, E. 1977: Folketingel under forandring. København: Samfundsvidenskabeligt
Forlag.

Damgaard, E. and Eliassen, K. A. 1975: 'Lovgivningsprocesser og politisk segmentering'
Århus: Institut for Statskundskab.

Damgaard, E. and Eliassen, K. A. 1976: 'Actors and Sectors in Danish Law-Making\ Paper
for ECPR-workshop, Louvain, 8 14 April, 1976.

Elvander, N. 1969: Intresseorganisationerna i dagens Sverige Lund: Gleerup Bokførlag.

Fröman, L. A. 1968: 'The Categorization of Policy Contents' pp. 41-52. Austin Ranney (ed.),
Political Science and Public Policy. Chicago: Markham Publ. Co.

Hague, D. C, Mackenzie W. J. M, and Barker A. (eds.) 1975: Public Policy and Private
Interests. The Institutions of Compromise. London: Macmillan.

Heisler, M. O. 1974: Patterns of European Politics: The 'European Polity Model'. In
Heisler (ed.), Politics in Europe. New York: David McKay Co.

Ingebrigtsen, 1968: Arnulf Ingebriglsen, De som styrer Norge, Oslo.

Jarlov, C, Johansen, L. N. and Kristensen, O. P. 1976: 'The Danish Committee System'.
Paper, IPSA World Congress, Edinburgh, August 16-21, 1976.

Jones, C. O. 1977: An Introduction to the Study of Public Policy. North Scituate, Mass.:
Duxbury Press.

Kvavik, R. 1976. Interest Groups in Norway. Oslo: Universitetsforlaget.

Lowi, T. J. 1964: 'American Business, Public Policy, Case Studies and Political Theory'.
World Politics 16, 677-715.

Meijer, H. M. 1969: 'Public Policy Formation in Sweden', Scandinavian Political Studies 4,
103-116.

Melander, K. A. 1974: 'Henvendelser til Stortinget 1919-1970'. Oslo: Institute of Political
Science.

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

Olsen, T. P. 1976: 'Organizational Participation in Government', University of Bergen.

Rokkan, S. 1966: 'Numerical Democracy and Corporate Pluralism', pp. 70-115 in Robert A.
Dahl (ed.), Political Oppositions in Western Democracies. New Haven: Yale University
Press

Ripley, R. B. and Franklin, G. A. 1976: Congress, the Bureaucracy, and Public Policy,
Homewood, 111.: The Dorsey Press.

Schmitter, P. C. 1974: 'Still the Century of Corporatism?' pp. 85-131. In Fredrick B. Pike &
Thomas Strich (eds.) The New Corporatism. University of Notre Dame Press.

Vogler, D. T. 1974: The Politics of Congress. Boston: Allyn & Cacon.