Resumé

The integration of organized interests into the formation and implementation of public policies is a core institutional trait of the Scandinavian countries. However, significant changes have taken place in the relations between organized interests and public authorities in Denmark and Norway during the last two decades. The use of traditional corporatist structures of interest intermediation has been reduced in favor of a corresponding increase in lobbyism. At the same time a marked increase in the frequency and intensity of contacts between organized interests and parliamentary actors has taken place. The shift in focus mirrors the increasing role played by the two parliaments in public policy formation and a less positive assessment of the outcomes of strongly institutionalized corporatist policy making by administrative decision makers.

Introduction 1

The integration of organized interests into public and administrative policy making is a core political structure of the Scandinavian countries. Numerous place the Scandinavian countries in the upper end on scales measuring the degree of integration of organized interests into public policy making (e.g., Lijphart & Crepaz 1991). In the Scandinavian countries, legislation has often been prepared on boards and in commissions or in other forms of close cooperation with organizations that represent the target groups of public policies. The implementation of public policies in many cases also involves organized interests as an integrated part of the implementation processes. However, significant changes in the position of organized interests in the policy making process seem to have taken place during the last decades. It is not yet clear, however, what exactly the changes are, how they should be explained, or what their consequences are.

---

* Peter Munk Christiansen, Department of Economics, Politics and Administration, Aalborg University, Fibigcrstræde 1, 9220 Aalborg, Denmark. E-mail: pmc@socsci.auc.dk; Hilmar Rommetvedt, Rogaland Research, Stavanger, and LOS Centre, Bergen. E-mail: Hilmar.Rommetvedt («)los.uib.no

The purpose of the article is two-fold. First, we describe some of the changes that have taken place in the position of organized interests in the policy making process in Denmark and Norway. We show that the relations between interest organizations and administrative actors are less institutionalized compared to 20 years ago, and we show that today parliamentary actors have more intensive contacts with interest organizations two decades ago. Second, departing in recent institutionalist theory, we delimit some of the explanations of the changing relations between interest organizations and parliaments.

Corporatism and Lobbyism

There are basically two kinds of institutions that govern the relations between organized interests and state actors, corporatism and lobbying. We define the corporatist channel of participation as an 'incorporation of interest groups into the process of policy formation and implementation' (Lijphart & Crepaz 1991, 235). As an administrative model, corporatism covers the participation of organized interests through membership of boards, councils, committees, etc., consultations, hearings, other formalized and often supported by a variety of informal contacts between and organized interests. Corporatism is, in other words, strongly institutionalized. When relations between organized interests and public authorities are mostly informal, ad hoc based, and the degree of institutionalization is low, we talk of lobbyism. Applied to real world phenomena, corporatism and lobbyism should be seen more as the two ends of a continuum than as discrete and distinct models. However, if we look at them as ideal types, we get the following pictures: Corporatism as an institution exactly specify which organizations enjoy privileged access to decision making arenas, whereas as the rules for access are much more open and ambiguous under lobbyism. The corporatist structure has to be set up by the authorities but once established, the initiative for action can be taken by interest organizations as well as by public authorities. Decisions are reached through negotiations with status quo as the default rule, and the actors involved in corporatist policy making can make de facto authoritative With lobbyism, the initiative primarily lies with interest organizations, the involvement of interest organizations in the decision making process is consultation on an ad hoc basis, and the relevant interest organizations are not directly involved in authoritative decision making.

The strategies of interest organizations can be directed towards bureaucraticas
as parliamentary actors. By combining the two dimensions,
institutionalization and direction, we get a fourfold typology as shown in

Figure 1 (cf. Rommetvedt & Opedal 1995). Since parliamentary corporatismbased
functional representation in legislature is not relevant in
contemporary Scandinavia, we exclude this category from further analysis.

Administrative and parliamentary lobbyism share some core characteristics. rules governing behavior tend to be ambiguous, and the degree of institutionalization is low. However, they are also very different, because they incorporate different actors which are under very different constraints. From the perspective of control and interest maximizers in interest organizations, institutions imply strong mutual commitment between the actors involved, whereas lobbyism is less demanding. Lobbyism is a more flexible institution with a greater potential for change than corporatism.

Corporatist Policy Making in Norway and Denmark

The strong integration of organized interests into political and administrative making in the Scandinavian countries started in the late 19th century and was expanded during the first half of the century (Nørgaard 1997). The strongly increasing public regulations of the private sector following two world wars and the economic crisis of the 1930s prepared the ground for the integration of organized interests into a large number of policy areas. By the end of World War 11, corporatist structures were well established in Denmark and Norway (Johansen & Kristensen 1982; Nordby 1994). The first three decades of postwar development proved the viability of the corporatist policy making system. Organizations came to hold a legitimate and institutionalized right to be involved in more or less all phases of political and administrative decision making (Olsen 1983,166ff.).

Johansen & Kristensen (1982, 197) find that issues that are non-ideological,measurable, narrow and technical' better facilitate corporatiststructures to issues with a heavy ideological load and issues on very general questions. Organizations basedon a functional

division of labor, i.e., business organizations and labor organizations, at the private as well as the public labor market, developed the strongest ties to the administration (Egeberg 1981; Johansen & Kristensen 1982; Olsen 1983; Nordby 1994).

The strongest means through which organizations were integrated were the numerous boards, committees, and councils established on a temporary basis in order to come up with proposals for solutions to small or large problems or on a permanent basis through which interest organizations became part of the administration through advisory or administrative roles. Boards and committees thus came to play a role on the input side of the political system as well as on the output side. Table 1 shows the developmentin number of boards, committees, etc. We can see that corporatist integration through these structures increased significantly from World War II and through the 19705. In Sweden, interest organizations even appointed members of the governing boards of the semi-independent state agencies (Olsen 1983, 167). A less constraining, but still important institutionwas norm that organizations were consulted through formal hearings in almost all proposals for legal and administrative rule making which could somehow be related to the interests of specific organizations. The integration of organized interests also took place on a less formalized, but not necessarily less important, basis. Interest organizations pursued a policy of creating and maintaining close contacts with civil servants and

ministers. Some of the large industrial and labor market organizations even
had daily contacts with civil servants at all levels in the central state administration(Buksti
Johansen & Kristensen 1982; Olsen 1983).

The development of strong corporatist structures apparently had consequences the role of parliaments in public policy making. In the mid-19605, Norwegian political scientist Stein Rokkan described the two channels of representation, 'numerical democracy' and 'corporate pluralism.' balance between the two channels was formulated in the phrase 'votes count, but resources decide' (Rokkan 1966, 105). It seemed to be a general observation that parliaments had come to play a smaller and smaller role in public policy making. Some quotations may give an impression the assessment of the relative strength of the two channels of participation: Robert B. Kvavik, an American scholar who studied the role of interest organizations in Norway, was 'surprised to discover the absence of "lobbyists"' in the Norwegian Parliament, Stortinget. He observed that 'legislation was shaped in the administration; once in parliament, the lines were fixed' (Kvavik 1976, 15, 120). Having interviewed the leaders of interest Kvavik concluded that '[parliamentary institutions receive an exceedingly weak evaluation' (ibid., 118). In the conclusion on the Norwegian power study, Hernes (1983, 303) wrote that there was agreement among the holders of power in the Norwegian society that power had been displaced 'from Stortinget towards the administration and the interest organizations.'

The parliaments in Norway and Denmark seemed to play only a modest role in public policy making. Apparently, these results were in close accordance with the literature on the 'decline of legislature' which dominated study of policy making in Western democracies in general and the study of parliaments in particular (Damgaard 1992a). To some extent, the observations made by students of corporatist participation on the limited role of parliaments in the Nordic countries are the result of a lack of scholarly interest in this role. When studying corporatism, one may be prone to look for signs of the strength of corporatism. However, to some extent, the observed weaknesses of parliaments in the policy process also reflect a real development.

Decline of Corporatism?

At the very same time as some of the above mentioned statements on the limited role of the Nordic parliaments were published, other scholars observedthat picture might look somewhat different. Johan P. Olsen, a colleague of Hernes in the Norwegian power study, in 1983 claimed that 'during the last part of the 1970s the Storting became a more rather than less

significant institution' (Olsen 1983, 72). In 1981, Egeberg (1981, 143) found that '. . . not only ministries and agencies, but also the cabinet and the institutions of Stortinget are core elements in the national environments of workers' and business organizations.' Damgaard (1992a) points to the increased activity of the Nordic parliaments in a number of respects. In 'The Strong Parliaments of Scandinavia,' he opposes the 'decline of legislature-thesis'directly 1994). Today, it is relevant to ask the opposite question of what was done in the 1960s and 19705: Is corporatism on decline?

In 1982, Johansen & Kristensen declared that 'public committees have long since become one of the most important, if not the single most important, mode of interest mediation' (p. 196). However, Table 1 shows that the use of boards, committees etc. has been reduced over the last decades. In Denmark, the number of committees peaked in 1980 as did the number of committees with members from interest organizations. The year registered as the peak year in Norway is 1977 for the total number of committees. The decline in the number of commissions, etc. during the 1980s and 1990s is significant in both countries. In Denmark, the proportion committees with members from interest organizations rose from 1980 through 1990 (73, 77, and 84 percent in 1980, 1985 and 1990) but fell significantly in 1995 (67 percent). However, the absolute number, which is what counts in real-life policy making, fell significantly. Further analysis on Danish data shows that the reduction in the number of committees is much more significant for committees dealing with the preparation of policy decisions than with the implementation of policies (Christiansen 1998). In Norway, the ratio of committees, etc. with members from interest organizations all committees did not change much after 1977, but the absolute number fell significantly. A parallel drop in the use of committees is also found in Sweden (SOU 1990).

If, as claimed by Johansen & Kristensen (1982), boards, committees,etc.are main structural mechanism that links public bureaucracy to organized interests, it is clear that this institution has come under strong pressure. However, it cannot be concluded that the use of boards, committees,etc. a linking mechanism is insignificant. First, there is still a significant number of committees with members from interest organizations.More analyses of Danish data show that the reduction in the number of committees is much less than average within ministries concerned with specific regulation, i.e., the traditional 'corporatist' ministries,such Agriculture, Industry, Labor, etc. (Christiansen 1994). Although a reduction is found in all types of ministries, the reduction is most significant in ministries responsible for public service production. This brings us to the second reservation. In Denmark and Norway, the public sector has become still more decentralized during the last two decades.

More and more responsibility as regards public sector management is taken care of at the local and regional levels. Third, the reduction in the number of committees etc. may, to some extent, also reflect the policies of the 1980s through which some reduction in the complexity of public sector organizationwas In some Danish ministries, a number of councils and committees were merged in order to reduce structural complexity without necessarily reducing the influence of interest organizations. Nordby (1994) correspondingly finds that many of the abolished Norwegian boards and committees were only of insignificant importance. Even with these reservations,we conclude that the most institutionalized element of traditionalcorporatism considerably less widespread in the 1990s than before.

Representation in commissions and boards has, to some extent, been replaced by contacts with bureaucracies on a lower level of institutionalization. 2 shows that even if the number of commissions and boards has been reduced, the total frequency of contacts between interest organizations administrative actors has increased from the beginning of the 1980s till the beginning of the 19905. It is interesting to observe that the least frequent type of contact ('Yearly') has increased more than the most frequent type of contact ('Monthly').

In order to control for the effect of the changing population of interest organizations between the two years observed, we have separated the organizations represented both years in the two national studies so that we are able to compare the development over time for the same group of organizations. The numbers for these organizations are shown in the tables in brackets. For all organizations, the increase in contacts is much larger in Denmark than in Norway. However, for the organizations represented in both years, the differences in contacts with the cabinet are of the same magnitude in the two countries, and for contacts to ministries and agencies the Danish organizations show a small drop in contacts and the Norwegian a small increase (Table 2, figures in parentheses).

In Table 3, we have split the organizations in two groups: The first group - business and labor organizations - are the traditional partners in corporatistinstitutions, the other group consists of all other organizations. The table shows that the increase in administrative contacts is primarily found for 'business and labor' organizations. In Denmark, the group of other organizations has small increases in administrative contacts except for frequent contacts with ministries and agencies which have dropped. In Norway, 'other organizations' contacts with cabinet and ministries are about the same in 1992 as in 1982, and contacts with agencies have dropped significantly, particularly the less frequent contacts. Again, the picture changes when we look at the organizations represented in both years. The Norwegian figures (in brackets) show that 'other' organizations have some

increase in less frequent contacts with cabinet, ministries and agencies comparedto where the figures for organizations represented both years are not that different from the figures for all organizations. In the Norwegian case, business and labor organizations represented both years generally have small increases in contacts with cabinet and ministries. Danish business and labor organizations represented both years have fewer contacts compared to all organizations.

In Denmark, there is also a significant increase in the proportion of organizations that assess contacts with administrative actors as important, cf. Table 4. (Since the Danish 1981 survey did not ask organizations to assess the importance of contacts, we have used data from 1976 in the Danish case.²) In Norway, there is no increase in the importance of administrative contacts. However, when we control for the change in the observed populations by looking only at organizations that are represented in both years, the two countries again converge (figures in parentheses). In Norway, we find a minor increase of five percent that find administrative contacts important. In Denmark, the overall 16 percent are reduced to ten percent for the organizations represented in both years.

In Table 5, the organizations are split into business and labor organizations other organizations. In Denmark, there is a strong tendency for an increasing part of business and labor organizations to say that their administrative contacts are important. This is also the case for 'other' organizations in Denmark, albeit less significant. In Norway, there is a certain increase in importance of administrative contacts for business and labor organizations, whereas 'other' Norwegian organizations show a small drop in the proportion that estimates administrative contacts to be important. we find that Danish and Norwegian organizations converge when we look only at organizations that were represented in both years (figures in parentheses). The increase in the assessment of the importance of administrative contacts is halved in the Danish case to 13 percent, and in the case of ministries turned from a two percent drop into an eight percent increase in the Norwegian case.

One should not push the conclusion too far: Corporatism is far from abolished, but use of the most institutionalized structure of traditional corporatism, boards, commissions, etc., is significantly lower today than 20 years ago. Although sectors with very strong corporatist structures remain, ministries are significantly less inclined to establish the traditional mediationstructures a problem needs to be solved. We must interpret the figures on the cluster of contacts as a kind of 'compensation' to this development.Contacts somewhat more frequent compared to a decade ago, they are assessed as significantly more important, and in particular the organizations that were earlier involved in strong corporatist structures tend to increase contacts and to assess their increasing importance. The

Danish development is more pronounced than the Norwegian. The decrease of formalized contacts through commissions and boards does not automaticallyimply corporatism is replaced by lobbyism. However, traditionalcorporatism under pressure, and the relative importance of administrative lobbyism is growing. With this conclusion, we turn to the relations between interest organizations and parliamentary actors.

Parliamentary Lobbyism in Denmark and Norway

Party groups and individual MPs have probably had contacts with interest organizations since modern parliaments were established. Some of the political parties have had, and still have, close relations to interest organizations. social democratic parties have always had close and even formalized relations with the labor movement, although the formal ties have been weakened through the last decade. The Danish Liberal Party and the Norwegian Center Party have had close contacts with agricultural organizations. However, these contacts have linked specific organizations and specific parties. Lobbyism is a broader phenomenon. Some elements of parliamentary lobbyism in which organizations have approached a number of parties and MPs have probably always existed in modern parliaments. If Kvavik's (1976) observation of the absence of lobbyism in the Norwegian parliament was correct, it is no longer true.

Table 2 shows that a quite substantial proportion of all nationwide interest organizations in Denmark and Norway have contacts with parliamentary 43 percent of Danish and 35 percent of Norwegian organizations contacts on at least a yearly basis, whereas the corresponding figures for monthly or more frequent contacts are 17 and 15 percent. Although the development is parallel in Denmark and Norway, the increase in contacts is much more outspoken in Denmark than in Norway. However, the difference in increases between Denmark and Norway is almost nonexistent we look only at organizations represented in both years in the two countries (figures in parentheses). Comparing all organizations with organizations represented both years within each country reveals that in the Danish case, the latter group has increased its contacts less than the former group, whereas it is the opposite in Norway.

If, again, we split the organizations into the two groups, business and labor organizations versus other organizations, as done in Table 3, we find a much stronger development for parliamentary contacts for business and labor organizations compared to the group of other organizations. In rough numbers, contacts on a yearly basis (and on a monthly basis in Denmark) doubled for business and labor organizations, whereas the development for other organizations is much more modest. Still, if we compare the group

of other organizations' contacts with parliamentary actors and the administration,these appear to have increased contacts to politicians,while contacts with the administration were almost constant. Parliamentary lobbyism is not first and foremost an activity reserved for organizations that confront problems of establishing institutionalized links with bureaucratic actors. Business organizations and labor organizations have by far the highest frequency of contacts with all types of actors. Again, we find that for all organizations the increases are much higher in Denmark than in Norway, whereas differences tend to be smaller when we compare the organizations represented in both years (figures in parentheses).

The importance of contacts between parliamentary actors and interest organizations also increased. Table 4 lists the proportion of organizations that found the contacts with parliamentary actors important. It shows that all types of contacts with parliamentary actors are assessed as more important in 1992/93 than in 1982/76. It also appears that the importance of parliamentary contacts has grown more than the importance of contacts with administrative actors. In the Norwegian case, there is, as mentioned above, even a small drop in the proportion of organizations that assess contacts with administrative actors as important. It should be noted that even if the importance of contacts with parliamentary actors has grown significantly, there are still more organizations that assess administrative contacts to be important than contacts with parliamentary actors. In the Danish case, the difference between the assessment of parliamentary and administrative contacts is much smaller than in the Norwegian case, and the number of organizations that assess political contacts to be important is generally twice as high in Denmark as in Norway.

If, again, we follow the organizations that responded both years (figures in parentheses), we find a convergence between the two countries. In the Danish case, these organizations assess political contacts to be more important the group of all organizations, but the increase in the proportion assessing contacts to be important is less for this group compared to all Danish organizations. In Norway, the same tendency for repetitive organizations found only in 1992, but not in 1982. As a consequence, the difference in percentages between 1976/1982 and 1993/1992 converges.

If organizations are split into the two groups also used above, we see from Table 5 that the proportion of business and labor organizations that find contacts with parliamentary actors important has risen significantly more than in the group of'other' organizations. In the Danish case, 58 percent of business and labor organizations find parliamentary contacts important in 1993 compared to 40 percent of 'other' organizations. The corresponding figures for Norway are 32 and 18 percent. Again, it should be noted that the score of administrative contacts is significantly higher for all groups in all years compared to parliamentary contacts. In both countries, we see that

differences between the assessment of political contacts between business and labor organizations, on the one hand, and the group of other organizations,on other hand, are insignificant in 1976/1982, but significant in 1993/1992. Since we found a corresponding cluster in the analysis of the frequency of contacts, we get another hint of a bias in the development of corporatist and lobbyist institutions in favor of the economic organizations.

As in the other cases, we find a convergence between Denmark and Norway when we restrict our analysis to the organizations represented in both years (figures in parentheses). In Denmark, the difference in the assessment the importance of contacts is +34 percent for all business and labor organizations for all contacts to parliament. The corresponding figure for Norway is +16. When we look only at repetitive organizations, the corresponding are +23 and +19. Differences are also converged when we look at the group of other organizations, with the difference that the Danish figures for repetitive organizations are not different from the figures for all other organizations.

In sum, there is clear evidence that contacts between interest organizations parliamentary actors have increased significantly from the early 1980s to the early 19905. Contacts are also considered more important today than in the early 1980s. There is still clear evidence of strong corporatist but parliamentary lobbyism has increased as traditional corporatism has declined in scope and intensity. The question to be addressed on the following pages relates to the second purpose of the article: How can these changes be explained?

Explaining Institutional Change and Inertia

The increased importance of administrative and parliamentary lobbyism may be the result of several and different developments, some of which are related to political and administrative institutions, and some of which are related to broader social and economic changes. We restrict our investigation the development of political and administrative institutions. Despite the vast scholarly and public interests in the relations between public authorities and organized interests, theories on these relations have not been strong in terms of explaining the changes, processes, and outcomes of corporatist institutions. Theories on corporatism (e.g., Schmitter 1974), policy networks (e.g. Rhodes & Marsh 1992a; 1992b), sectorization and segmentation (Egeberg et al. 1978) suffer from deficiencies in their explanatory (Heisler 1979). They tend to emphasize description in favor of explanation. In addition, and important in the context of this article, they tend to ignore the role played by parliaments.³ We cannot overcome these

problems in this article. However, departing in recent institutionalist
theory, we can throw some light on some elements of the essentials and
dynamics of the relations between state and organized interests.
Institutional theory has developed extensively during the last decade,

and many theoretical paths have been developed. A basic claim is that political, economic, and social behavior cannot be understood without looking at the relations between actors and the institutions of political, economic, and social life. Institutions affect the outcome of social interactions. major distinction is whether actors are considered primarily calculating or primarily rule followers (Marsh & Olsen 1989). Rational theory departs in rational choice theory and assumes that actors are selfinterested intendedly rational - although limited by cognitive capacity and uncertainty. For real world analysis, the demands on rationality need not be very restrictive. Ostrom assumes that actors are calculating. Actors are expected to '. . . compare expected benefits and costs of action prior to adopting strategies for action' (Ostrom 1991, 243).

Institutions are defined as rules and norms that guide individual behavior by prescribing, prohibiting, or encouraging courses of action. Institutions are constraints on individual behavior as well as simultaneously providing individuals predictability and stability. Institutions are objects of strategic actions from self-interested actors (Ostrom 1986; 1991; Shepsle 1989; North 1990).

Institutions reflect the power of the actors who established them (North 1990; Knight 1992). Powerful actors may choose to change institutions if they have the power to do so and if they calculate change to be in their favor. Consequently, institutions are biased in favor of some actors and in the disfavor of others (Knight 1992, passim; North 1990, 48).

Our choice to draw on the rational variants as a frame for understanding the interrelations between organized interests and public authorities is based on two presumptions: First, from the outset it seems reasonable to assume that interest organizations are calculating actors.⁴ They have to deploy their resources in the best way in order to maximize benefits. Second, rational theories focus more explicitly on power and interests compared to the sociological variants. The interrelations between organized interests and bureaucracies and parliaments are basically about interests and power.

The preferences of politicians, bureaucrats, and representatives of interest undoubtedly vary significantly. Yet, when we restrict our considerations to the interplay between public authorities and organized we might come up with a set of preferences that are not all that different among the different types of actors.

First, the actors involved in institutionalized interrelations between
organized interests and state actors have many and complex preferences,

some of which contradict each other. In a complex world, actors may not even be able to fully range preferences in order to satisfy the demand for transitive preferences. As mentioned, we relax the assumptions made in neo-classic theory. For the sake of simplicity, we assume that political and bureaucratic actors and actors that represent organized interests are control and authority maximizers (Christiansen 1996; Nørgaard 1997, 14), which means that actors try to acquire knowledge about the activities and strategiesof actors, to influence the strategies of other actors, and to influence the outcomes of decisions. In strongly institutionalized settings, actors will maximize long-term influence, and may thus accept short-term sub-optimal outcomes or even losses if they calculate this to be in their long-term interest. Since institutions reduce uncertainty as regards the social behavior of others (Knight 1992), involvement in institutionalized mutual relations minimizes the complexity and uncertainty stemming from other actors.

For bureaucrats, strongly institutionalized relations with interest organizations the possibility of parliamentary acceptance and successful implementation of new policies. They may also be exploited when bureaucrats ministers are in conflict with other ministries in the pursuit of institutional interests (Christiansen 1996). Parliamentary actors may have incentives to institutionalize their relation with organized interests if they fit other interests and preferred strategies. However, it is important to notice that the institutional environments of bureaucrats and politicians may conflict with the development of strongly institutionalized relations with organized interests. This is more the case for politicians than for bureaucrats, the former are constrained by parliamentarism and the necessity to build coalitions with other parties. As a consequence, it is less likely, compared to bureaucratic actors, that parliamentary actors will build strongly institutionalized relations with organized interests. Some oT the political parties (e.g., labor and agrarian parties) traditionally have strong ties with labor and farmer unions. However, these ties seem to have loosened during the last decade, and in any case they have to be weighed against the constraints stemming from alliances with other parties.

Second, for representatives of interest organizations, integration in the political and administrative decision making process is a means to create, maintain or increase substantial benefits. Integration in public policy making a privileged position which allows some actors to pursue the substantial goals of an organization. Substantial goals are many things, such as subsidies, avoiding taxation, influencing public regulation with a specific substantial position, protecting a monopoly position, etc., etc.

Institutional theory most often have problems with explaining change.
However, Knight (1992) provides a sound theory of change in a rationalist
institutionalist perspective. The premises are that calculating actors are in

pursuit of distributive advantages and that institutions reflect the power of the actors who established them. Biased institutions will always confront a real or potential threat of pressure for change, but change is only likely to occur when strong actors are in favor of change and when they are able to agree with or force change on other actors.

Knight (1992, 145ff, 171ff.) suggests that one or both of the following two conditions are necessary for institutional change to occur: (1) the relative of actors change, and/or (2) institutions produce outcomes that are no longer desirable for powerful actors. The conditions for institutional change are restrictive, since institutional change may fail to occur when actors choose to comply with existing institutions because defection is estimated to be worse. A reluctance towards institutional change due to the uncertainty connected with the establishment of new rules also points to delays in existing institutions.

As control and influence maximizers, interest organizations deploy their resources where they assess the largest payoff in their pursuit of control and influence. As long as they get maximum control over the decision making process through well-known corporatist institutions, they prefer to continue to do so. If, for one or another reason, access to traditional corporatist institutions restricted, interest organizations will exploit other channels of influence, such as administrative lobbyism. Similarly, interest organizations respond to relative shifts in power between bureaucracy and parliament by redirecting their attempts to influence policy making towards the more powerful authority.

Changes in the Political Conditions for Corporatism and Lobbyism

For many years, the Danish Folketing and the Norwegian Storting appeared to be stable and consensual parliaments. Even if the two political systems are both multi-party systems, they proved to establish stable governments in the first decades after World War 11. In terms of parliamentary majority, Norway had stronger governments than Denmark which has primarily had minority governments since World War II (Damgaard 1992b; Rommetvedt 1992). However, significant changes occurred in the 19705.

The most dramatic changes in the Danish parliament happened in 1973, when the traditional party system, which had only changed slightly since the early 19205, more or less broke down. The number of parties in Folketinget doubled, and during the two following decades the conditions for creating majorities in parliament significantly worsened, with frequent elections as result. During the Conservative-Liberal governments in the 1980s, parliamentary norms changed so that it became accepted by governmentsto

mentstostay in power while accepting defeats in parliament. In a number of cases, the Danish governments had to administer policies decided upon in parliament by the opposition. As a consequence, the outcome of parliamentarydecision processes is significantly less predictable than before 1973.

Alongside this development, institutional reforms and increased activity in Folketinget increased the relative power of Folketinget vis-å-vis the administration. The establishment in the early 1970s of permanent committees to mirror the division of labor between ministries increased the capacity of MPs and committees to follow and control the performance of ministers (Damgaard 1977). The activities of the committees sharply during the 19705. Also plenary activities rose in terms of the number of interpellations and the number of questions from MPs to ministers.

The Norwegian development is, in many respects, parallel to the Danish experiences. New parties also entered Stortinget in 1973. The previously established, stable traits of parliamentary governance were weakened. The tendency for recommendations from parliamentary committees to be accepted was weakened. Before 1973, dissent only occurred in 16 percent of the committee recommendations (budget not included) compared 35 percent between 1981-89, 59 percent between 1989-93, and as high as 67 percent in 1993-94. The Norwegian cabinets also confronted non-cabinet majorities in Stortinget during the 1980s and 19905. On average, the government lost less than one vote per month in 1979-81, 4.6 votes per month in 1981-83, and 2.7 per month in 1983-86. The corresponding for the two periods 1986-89 and 1990-95 were 7.2 and 7.4 votes per month respectively (Rommetvedt 1998).

Stortinget has had standing committees for many years, however committees party groups in particular have experienced a significant increase administrative resources and staff (Rommetvedt 1995; 1998). The above-mentioned figures on the number of dissents point to an increased level of bargaining activities in parliament. Private member bills are increasingly introduced. In 1984/85, 15 such bills were proposed compared with 113 in 1995/96 (Rommetvedt 1997, 103).

In sum, if we look at the political and administrative system from the viewpoint of interest organizations in pursuit of control and influence, the institutional conditions of success have changed significantly during the last 25 years and particularly since the early 1980s. The level of conflict and politicization in the two parliaments has increased dramatically. At the same time, the capacity of parties, parliamentary committees and MPs to control the government and the individual ministers has increased. The pre-1970period relatively stable governments, a relatively low level of conflict and politicization, and parliamentary accept of the administration

as the owner of the rights of initiative and preparation of legal regulations is long gone. The relative power of the parliaments vis-å-vis ministers and administration has changed in favor of the former. The outcome of parliamentaryactivity become much less predictable, and the probability that governmental proposals will be changed significantly through parliamentarytreatment quite high.

Under such conditions, rational actors will be more attentive to parliamentary in their pursuit of control and influence. They have much stronger incentives to be aware of what is going on in Parliament and to let parliamentary committees, party groups, and MPs know about their general and specific policy preferences. Although the institution of parliamentary is much more ambiguous, and thus difficult to act within, rational organizations have little choice but to increase their attention towards and contacts with parliamentary actors. Since it is much more difficult to control processes and outcomes through parliamentary lobbyism than through corporatism, organizations may prefer the stable and predictable of corporatist institutions if they can pursue their goals through corporatist participation. However, due to the relative power shift between parliaments and administration, a strategy directed towards administrative actors only is dangerous. In addition, lobbyism is a more flexible strategy. Lobbyism is not dependent on the willingness of political authorities to establish corporatist structures.

If we change our perspective and look at corporatism versus lobbyism from the viewpoint of parties, MPs, and ministers, we find that some of the advantages of lobbyism over corporatism have become more important over the last decades. The rising level of activity and conflict in the Danish and Norwegian parliaments may be related to the increased volatility of voters. Eighteen percent of the Norwegian voters who voted at both elections 1965 and 1969 changed party (Valen 1981, 335). From 1993 to 1997, the corresponding figure was 33 percent (Statistisk sentralbyrå 1998). Measured in terms of the average net electoral gains by the political parties, voter volatility increased from 7.1 percent in the period 1950-69 to 13.5 percent in 1970-95 in Denmark. The corresponding figures for Norway are 4.3 and 13.8 percent (Andersen & Hoff 1998). The higher voter volatility implies an intensified competition between political parties. MPs and parties, on the one hand, have to demonstrate their political abilities in the mass media. Critical journalists, on the other hand, demand a statement or an initiative in the continuous stream of new burning issues on the political agenda. Issues and conflicts compete for attention. The ever increasing pace of modern politics necessitates flexibility and swift adaptation to new challenges.

The demand on ministers may even be greater than on parliamentary
actors. Ministers are in charge of large, professionalized ministries which

possess the capacity to take action in order to solve small and large problems.The as a consequence, expects minister and ministry to take due action when necessary. At the same time, ministers possess the formal authority to decide if, how, and when interest organizations should be integrated into the preparation and implementation of new or revised policies. In the competition for visibility and attention, the establishment of a public committee may not be viewed as a sign of willingness to solve the problem in a foreseeable future. Ministers can hardly avoid inviting interest organizations to let their voice be heard. But they may have good reasons to prefer other types of integration than the strongly institutionalized public committees.

As a consequence of this development, stability fostered by corporatist institutions becomes less important and, in some cases, even an obstacle for the pursuit of political strategies. Politicians need more leeway for their own initiatives. They are less willing to commit themselves to policies worked out by bureaucrats and representatives of interest organizations in corporatist negotiations. Politicians would also like to have a foot in different camps and to draw on the expertise of various organizations. In other words, a more flexible and less committing form of contact with organized interest, i.e., lobbyism, is conducive for the modern politician as well.

In sum, it appears to be plausible to explain the changes in corporatist interest intermediation in Denmark and Norway at least partly by the increased power of parliaments vis-å-vis the executives and a less intense preference for the outcome of corporatist policy making.

Conclusions

Institutions that integrate organized interests into political and administrativedecision are the core institutions of a democratic society. With the significantly broadened scope and intensity of political interference into a large number of societal activities, one can hardly imagine that the electoral-parliamentary channel alone can cope with the demands on the political system. In Lewin's (1992) phrase, society has to somehow solve the 'intensity problem' of democratic societies. Minorities with intense interests in a policy question risk being outvoted by the majority with less intense interests if all decisions have to go through the parliamentary channel of participation. As a consequence, political and administrative actors must somehow intermediate the narrow interests of interest organizations with broader societal interests. According to Lewin, corporatism is the solution to the 'intensity problem.' Recently, however, according to Lewin, corporatismhas

poratismhasperverted itself because it grew into a general channel of
excessive demands on the political system (ibid., 128ff.).

Post-war Scandinavian corporatism became a very strong institution that may have confirmed the worries of some observers of the 1960s and 19705. In many policy areas, it became accepted that the right to initiate, formulate, implement policies was placed in administrative and corporatist arenas. This gave strong interest organizations a very significant position in considerations on future policies. In pursuit of control and influence, strong organized interests succeeded in acquiring a large say, sometimes even a de facto veto right, in policy making. The establishment of a large number of boards, commissions, committees, etc. and very intense contacts with bureaucratic actors were the institutions through which organizations were integrated in public policy making.

Corporatism may be an adequate solution to the intensity problem in relatively homogenous societies. Compared to international standards, Denmark Norway are homogenous countries. In recent years, however, considerable changes have taken place. The two countries are not as unitary as they used to be. The party systems have become more fragmented, and the number of interest organizations has increased. In most areas, there are several different interest organizations with equally intense, but conflicting, interests trying to influence the policy making processes. In short, Denmark and Norway have become less unitary and more pluralist societies than before. As a consequence, corporatist negotiations get complicated. It is difficult to set up an adequate corporatist structure to handle the problems when the number of relevant actors is high. Consequently, less formal and less institutionalized forms of participation, i.e., lobbyism, may turn out to be the alternative.

For a number of reasons, the Danish and Norwegian parliaments increased power vis-å-vis the administration. The tradition of stability and consensual policy making to some extent eroded during the 1970s and 1980s. Interest organizations could no longer be sure that the 'normal' way of enacting legislation was through careful and detailed preparation in corporatist structures and formal confirmation in parliament. As a consequence, control and influence seeking interest organizations had to devote more attention to parliamentary activity, including the establishment regular contacts with parliamentary committees, party groups, and MPs. At the same time, the outcomes of corporatist policy making became less attractive to ministers and top civil servants. Increasing party competition, competition for attention in the media, the 'public' demand for due action in response to policy problems, and the speed of the policy making process today are less compatible with the existence of strongly institutionalized relations with interest organizations compared to the conditions the policy making process 20 years ago.

The theoretical preconditions of change of well-established institutions are demanding. However, they are present in relation to the Danish and Norwegian parliaments, which have the prerogative to enact legislation. Even if interest organizations would prefer the stable structures of traditional to the much more unpredictable and ambiguous institution parliamentary lobbyism, they cannot choose if and when parliament increases its power vis-å-vis the administration. Similarly, when the outcomes of strong corporatist structures are assessed to be less attractive for ministers and top civil servants, they have the power, even if that power is not unrestricted, to force upon interest organizations elements of new ways of integrating interest organizations into the administrative decision making process.

The comparative approach pursued in this article has focused on parallel developments in Denmark and Norway as regards the relations between state actors and interest organizations. The fact that these relations have developed so similarly in the two countries proves that they are not only to be explained by idiosyncratic attributes of the two countries, but that we are dealing with phenomena shared with other political systems comparable to those of Denmark and Norway. Actually, we have found similar changes in as different countries as Brazil and Austria (Boschi 1997; Crepaz 1994). More detailed studies of policy sectors probably would enlighten significant differences between the two countries in the way organized interests are integrated into the political and administrative decision making processes. However, knowledge of the general development in the relations between state and organized interests in the two countries would be of significant value also for sectoral studies.

Appendix: Data

1. Danish data on boards, councils, commissions etc.: Unlike Norway, Denmark has no official register of public boards, commissions, etc. Data have been collected by different scholars through the last two decades from a large number of sources. The period 1946 through 1975 is documented in Kristensen (1979). Data from the period 1980 through 1990 have been published in Jørgen Grønnegård Christensen & Peter Munk Christiansen, 1992, Forvaltning og omgivelser. Herning: Systime (pp. 66ff). Data for 1995 have recently been collected at the Department of Political Science, University of Aarhus. Documentation for data from 1980 through 1995 is available in Jørgen Grønnegård Christensen & Peter Munk Christiansen, Kodebog for udvalgsdatabase, Aarhus: Department of Political Science.

2. Norwegian data on boards, councils, commissions etc.: Data are collected by Planlegging- og Samordnings-departementet and available from The Norwegian Social Science Data Services. Data are published every fourth year in StortingsmeldingNo. and in Oversikt over slatlige udvalg og råd m.v., Oslo: PlanleggingsogSamordnings-departementet.

ogSamordnings-departementet.Data have been made available by John-Erik
Ågotnes, Norwegian Social Science Data Services.

3. Danish survey data on contacts between interest organizations and public authorities: have been collected through three questionnaire surveys of all national Danish interest organizations by scholars at the Department of Political Science, University of Aarhus. The 1976 data are available from DDA-0053 (1976), Interesseorganisationer Danmark 1976, Odense: Danish Data Archives and 1981 data from DDA-0510 (1981), Interesseorganisationer i Danmark 1981, Odense: Danish Data Archives. DDA-0053 (1976). 1993 data are from the project Interesseorganisationernes struktur, opgaver og kontakt til myndigheder, 1993. The survey was carried out by Jørgen Grønnegård Christensen, Peter Munk Christiansen, and Niels Chr. Sidenius, Department of Political Science, University of Aarhus. The population was delimited in the same way as in the previous surveys. The population is delimited from a large number of sources. As a consequence, it is difficult to precisely the total population. The theoretical criterion is all national Danish interest organizations, excluding pure hobby or leisure organizations. The population was 2050 organizations, and 1316 organizations responded. The only known bias in the responses is that a number of industrial organizations have not responded. The Danish Manufacturing Employers Association did not want to participate in the survey on behalf of the organizations for which it provides secretarial assistance. Nevertheless, we received answers from a number of these organizations.

4. Norwegian survey data on contacts between interest organizations and public authorities: Data have been collected through questionnaire surveys of all national Norwegian interest organizations by the Department of Political Science, University of Oslo and by the University of Tromsø. Data for the years 1964, 1967, 1971, 1976, 1983, and 1992 are available at the Norwegian Social Science Data Services. The population of organizations is delimited in a way quite similar to the Danish surveys, although there may be some differences in the selection procedures. Since the population is delimited from a large number of sources, it is difficult to precisely define the total population. The surveys are documented in Martin-Arne Andersen & Jan-Ståle Lauritsen, Organisationsarkivet. Ei brukarrettleiing, NSD Report No. 86, Bergen: The Norwegian Social Science Data Services and in Abraham Hallenstvedt & Jon Trollvik, eds., 1993, Norske organisasjoner, Oslo: Fabritius. Data have been analyzed and made available to us by Endre Holmefjord, 1998. Organisasjonene, forvaltningen og de folkevalgte. Report R9BOB. Bergen: LOS-senteret.

---

NOTES

1. An earlier version of this article was presented at the XVII IPSA World Congress, Seoul, 17-21 August 1997. The authors would like to thank Endre Holmefjord and Kirsten Nielsen for their assistance.

2. Due to differences in the alternatives the respondents were presented with, crosscountry of the importance of contacts should be done carefully, cf. notes for Table 4 and 5.

3. Egeberget al. (1978) include members of parliamentary committees in their description of political segments.

4. Rational institutional theory adheres to the principle of methodological individualism, which means that the unit of analysis is, in principle, individuals, and that aggregate actors are, in principle, dissolved to the level of individual actors. Organized interests are, per definition, collective actors. As a consequence, we must assume that we can treat interest organizations as Unitarian actors. According to Scharpf, this is not a prohibitive assumption. He concludes that 'the prevailing practice of treating a plurality of actors as a single player in game-theoretical applications is often fully justified' (Scharpf 1991, 291). In empirical analysis, it is from time to time clear that the assumption is not justified. In such cases, we must deal explicitly with the problem and try to disaggregate collective actors.

4. Rational institutional theory adheres to the principle of methodological individualism, which means that the unit of analysis is, in principle, individuals, and that aggregate actors are, in principle, dissolved to the level of individual actors. Organized interests are, per definition, collective actors. As a consequence, we must assume that we can treat interest organizations as Unitarian actors. According to Scharpf, this is not a prohibitive assumption. He concludes that 'the prevailing practice of treating a plurality of actors as a single player in game-theoretical applications is often fully justified' (Scharpf 1991, 291). In empirical analysis, it is from time to time clear that the assumption is not justified. In such cases, we must deal explicitly with the problem and try to disaggregate collective actors.

REFERENCES

Andersen, J. G. & J. Hoff. 1998. Democratic citizenship in the Scandinavian Welfare States,
Aalborg: Aalborg University (unpublished manus.).

Boschi, R. 1997. The Corporatist Heritage and New Patterns of Interest Representation in
Brazil. Paper presented at the XVII IPSA World Congress. Seoul. 17-21 August.

Buksti, J. 1985. Organisationernes folk. Aarhus: Forlaget Politica.

Christiansen, P. M. 1994. 'Organisationernes rolle: Det gode samfund eller organiseret
tyranni?,' in Ibsen, M., Nielsen, P. L. & Schou, 8., eds., Folkestyre og Forvaltning. Copenhagen:
og Økonomforbundets Forlag.

Christiansen, P. M. 1996. Why are Corporatist Arrangements so Attractive? State and
Organized Interests in a Rationalist Institutionalist Perspective. Aarhus: Department of
Political Science.

Christiansen. P. M. 1998. 'Interesseorganisationer, centraladministration og udviklingen af nye
indflydelsesstrategier', in Ronit, K., ed., Interesseorganisationer i dansk politik. Copenhagen:
Jurist- og Økonomforbundets Forlag.

Crepaz, M. M. L. 1994. 'From Semisovereignty to Soverereignty. The Decline of Corporatism
and Rise of Parliament in Austria,' Comparative Politics 27. 45-65.

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

Damgaard, E. 1992a. 'Nordic Parliamentary Government,' in Damgaard, E., ed., Parliamentary
in the Nordic Countries. Oslo: Scandinavian University Press.

Damgaard, E. 1992b. 'Parliamentary Change in the Nordic Countries,' in Damgaard, E.,
ed., Parliamentary Change in the Nordic Countries. Oslo: Scandinavian University Press.

Damgaard, E. 1994. 'The Strong Parliaments of Scandinavia.' in Copeland. G. W. &
Patterson, S. C, eds., Parliaments in the Modem World. Ann Arbor: The University of
Michigan Press.

Egeberg, M., Olsen, J. P. & Sætren, H. 1978. 'Organisasjonssamfunnet og den segmenterte
stat,' in Olsen, J. P., ed.. Politisk organisering. Bergen: Universitetsforlaget.

Egeberg, M. 1981. Stat og organisationer. Bergen: Universitetsforlaget.

Heisler, M. O. 1979. 'Corporate Pluralism Revisited: Where is the Theory?.' Scandinavian
Political Studies 2,277-97.

Hernes, G. 1983. Det moderne Norge. Bind 5: Makl og styring. Oslo: Gyldendal.

Johansen. L. N. & Kristensen, O. P. 1982. 'Corporatist Traits in Denmark, 1946-1976/ in
Lehmbruch, G. & Schmitter, P., eds.. Patterns of Corporatist Policy-Making. Beverly Hills:
Sage.

Knight, J. 1992. Institutions and Social Conflict. Cambndge: Cambridge University Press.

Kvavik, R. B. 1976. Interest Groups in Norwegian Politics, Oslo: Universitetsforlaget.

Lijphart, A. & Crepaz, M. M. L. 1991. 'Corporatism and Consensus Democracy in Eighteen
Countries: Conceptual and Empirical Linkages,' British Journal of Political Science 21,
235 56.

March. J. & Olsen. J. P. 1989. Rediscovering Institutions The Organizational Basis of Politics.
New York: The Frec Press.

Nordby, T. 1994. Korporatismepå norsk. 1920-1990. Oslo: Universitetsforlaget.

Nørgaard, A. S. 1997. The Politics of Institutional Control. Corporatism in Danish Occupational
and Health Regulation & Unemployment Insurance. 1870-1995. Aarhus:
Politica.

Olsen, J. P. 1983. Organized Democracy. Political Institutions in a Welfare State - The Case
of Norway. Oslo: Universitetsforlaget.

Ostrom, E. 1986. 'An Agenda for the Study of Institutions,' Public Choice 48, 3-25.

Ostrom, E. 1991. 'Rational Choice Theory and Institutional Analysis: Toward Complementarity,'
Political Science Review 85, 237-43.

Rhodes, R. A. W. & Marsh, D. 1992a. 'Policy Networks in British Politics. A Critique of
Existing Approaches,' Marsh, D. & Rhodes, R. A. W., eds., Policy Networks in British
Government. Oxford: Clarendon Press.

Rhodes, R. A. W. & Marsh, D. 1992b. 'Policy Communities and Issue Networks. Beyond
Typologi,' in Marsh, D. & Rhodes, R. A. W., eds., Policy Networks in British Government.
Oxford: Clarendon Press.

Rokkan, S. 1966. 'Norway: Numerical Democracy and Corporate Pluralism,' in Dahl, R.
A.. ed., Political Oppositions in Western Democracies. New Haven: Yale University Press.

Rommetvedt, H. 1995. 'Personellressurser, aktivitetsnivå og innflytelsesmuligheter i et
Storting i vekst,' Norsk Statsvitenskapelig Tidsskrift 11, 251-76.

Rommetvedt, H. 1997. 'Næringslivets politiske påvirkning,' in Malmø, M., ed., Politisk
Årbok 1997. Oslo: Andante Forlag.

Rommetvedt, H. 1998. 'Norwegian Parliamentary Committees: Performance, Structural
Change, and External Relations,' Journal of Legislative Studies 4:1.

Rommetvedt, H. & Opedal, S. 1995. 'Miljølobbyisme og næringskorporatisme?,' Nordisk
Administrativt Tidsskrift 76, 279-302.

Scharpf, F. W. 1991. 'Games Real Actors Could Play: The Challenge of Complexity,' Journal
of Theoretical Politics 3, 277-304.

Schmitter, P. C. 1974. 'Still the Century of Corporatism?,' Review of Politics 36, 85-131.

SOU 1990. Statens ojfentliga utredningar. Demokrati och makt i Sverige. Maktutredningens
huvudrapport. Stockholm.

Statistisk sentralbyrå. 1998. 'Valgundersøkelsen, 1997,' in Ukens statistikk 6.