Modern democracy combines two forms of governance. One is representative based as it is on the principles of universal elections, parliamentary representation, and a government in charge of governmental business. One way or another, this government is accountable to the representative body and ultimately to the electorate. The other form of governance is corporatist. It is based on the idea that individuals whose interests are affected by similar conditions organize themselves to make their specific demands count politically.

Through either of these two systems, citizens can present their demands for or against political action. Through them, policy-makers can mobilize support for political action and eventually seek to have their decisions on public policy legitimized. But when it comes to the implementation of policy, only theories of representative government give any indication as to how this is to be organized, how policy implementation is to be related to policy and policy-making, and how the administrative executors of policy are to be held accountable for their decisions.

The vision is simple and beautiful. The organization responsible for policy
implementation is organized along bureaucratic principles. Its structure is

Resumé

Danish democracy has strong corporatist characteristics. The implication is that interest organizations are integrated in both policy-making and policy implementation. This article deals with policy implementation. A distinction between different administrative regimes is developed. The proposition advanced here is that variations with respect to the involvement of interest organizations in administrative execution of public policy have consequences for administrative behavior. A number of scandals revealed during recent years concerning the management of public funds provide the basis of the empirical analysis. The conclusion is that with increased involvement of interest organizations in the administrative implementation of policy, legal norms are superseded by political norms, while the traditional institutions of administrative accountability are weakened.

hierarchical. Competences are allocated in a way that does not leave open any doubt as to their precise contents and as to which civil servants are responsible for doing what. The clarity in structure and competences is further stressed through the presumed professionalism of the bureaucratic staff. Civil servants recruited solely on their professional merits serve the organization in life-long careers. These civil servants are subjected to guidance provided by their superior within the organization - in governmentalorganizations is ultimately the minister who is the political leader of the organization. But in Max Weber's words, the loyalty of the civil servants is directed "towards the rules, not towards the person" (Weber 1922, 549-550). By stressing the obligation of civil servants to obey general rules laid down by law and by incorporating civil servants in bureaucratic organizations, Weber argued that it was possible for citizens to calculate the decisions of the civil service. This predictability of bureaucratic behavior is based on the assumption that civil servants will not bias their decisions through considerations which are not authorized and legitimized by law (Weber 1922,562-563). According to Weber, "the fully developed bureaucracypays to the principle of 'sine ira ac studio'" (Weber 1922, 563).

This, according to the theory, is also clear in other respects. Thanks to the way it has been organized, bureaucracy is the type of organization which is comparatively most reliable when it comes to paying respect to the values listed above. Although it is long recognized that bureaucracy will often accumulate considerable power, this problem can be minimized. A high degree of openness as to its operation is one strategy; another strategy is to subject bureaucracy to strict external control, e.g. through oversight by parliamentary committees, critical auditing, judicial oversight through the courts and ombudsmen (Day and Klein 1987).

The theory of corporatist democracy, by comparison, refers to "the incorporation of interest groups into the process of policy formation and implementation" (Lijphart & Crepaz 1991, 235). Because of its origin, this body of theories has not developed a similarly coherent view of the relationship between policy-making and bureaucratic implementation. Actually, although these theories by definition stress the corporatist aspects of both policy-making and policy implementation, most theoretical and empirical research has focused on analyzing corporatist policy-making. This leaves open the issue of corporatist participation in and influence on policy implementation and its consequences. This is the subject of the current article.

In the remainder of the article, attention is devoted to the nature of corporatist democracy and its implications for the implementation of public policy. First, a typology of administrative regimes is developed to describe the forms and degree of interest group involvement in policy implementation.The is that the character of the administrative regime

has implications for administrative behavior and for administrative accountability.Then, framework is applied in an analysis of a number of public scandals which have been revealed in Denmark since the late 19705. In conclusion, the article discusses the need for a more precise analysis of corporatism. It is argued that not only delegated administration but also the administrative regime - here defined as corporatist bureaucracy - are promising fields of research.

Corporatism and the Implementation of Public Policy

One possibility is that interest organizations have no direct influence on the implementation of public policy. They might stop with their efforts to influence policy-making, typically the contents of laws and regulations or the allocation of public funds for purposes supported by them. In this case, implementation could take place within the framework described above. Civil servants would concentrate on their jobs as professional and impartial executors of political decisions. Any problems that might arise could still be addressed through the institutional devices supplied by the theory of a democratically accountable bureaucracy.

Interest organizations cannot be expected to stop their efforts to influence public policy at the gates of public bureaucracy, however. Experience tells them that policy phrased in the general terms of a statute and annual budgetary allocations does not determine the contents of policy delivered to their membership. This gives them a strong incentive to pay an active interest not only in policy implementation, but also in the institutional framework within which policy implementation is staged (Moe 1989, 269-277). interest organization executives ask themselves: How can we make sure that general policy isn't implemented to our disadvantage? And they add the equally pertinent question: Are there ways in which we could make sure that policy is actually implemented to our advantage?

In this situation, organization executives are confronted with a strategic choice. In Table 1, a distinction is made between four different situations. They differ as to the distribution of competences between public bureaucracyon one hand and interest organizations on the other. According to traditional theories of representative government, there is no room for interest organizations exerting their direct influence on administrative decision-making. In this case of pure bureaucracy, civil servants make decisions which are subsumed under the general rules regulating the specific case under consideration. Interest organizations are kept at a distance from decision-making. Their only possibility of influencing administrative

implementation is at the preceding stage of policy-making where they may press for influence. A supplementary strategy could be to try to influence the allocation of administrative competences within government, e.g. through demands for having their specialized department or agency (Christensen 1980, 47-50).

In a more ambitious strategy, interest organizations would opt for access to influencing administrative decision-making without questioning the formal of bureaucracy to make the final decisions on how policy should be implemented. In this case of consultative bureaucracy, interest organizations acquire the right to make themselves heard on general or specific matters of administrative decision-making. This advisory capacity of interest organizations may be due to a totally informal status. But interest organizations will probably try to have it formalized through the establishment of advisory committees and working parties. Here on a regular basis they can expect to be consulted by the relevant branch of the civil service. Their preference is to have the relevant statute phrased in a way which de lege guarantees their permanent participation in administrative decision-making. This kind of consultative bureaucracy constitutes an important aspect of Danish public administration (Bent Christensen 1958; Kristensen 1979; Christensen & Christiansen 1992, 71-78).

Interest organizations, however, have no reason for renouncing claims for formal co-responsibility of policy implementation. By setting up a corporatist bureaucracy, the civil service is politically committed to share its administrative responsibilities with interest representatives. Often top civil servants and organization leaders will share a joint interest in this kind of corporatist bureaucracy (Christensen & Christiansen 1992, 8-13). In Danish public administration, public bureaucracy is corporatized in several ways. Implementation of regulatory policy will often be based on decisions

made on boards and committees on which both policy responsible governmentdepartments interest organizations are represented. In some cases these corporatist bodies take over administrative decision-making, leaving only the paper work to the civil service. In other cases, corporatist bodies act as committees of appeal to which citizens and private firms can bring complaints over decisions formally made within the civil service. Danish interest organizations have shown considerable effectiveness when it comes to defending this particular form of corporatist bureaucracy, reserving a right of appeal to their membership and giving them a final say in the application of policy in specific cases (Christensen 1991, 152-153 and 172-182).

Responsibility for many public services in Denmark is placed with public sector organizations that employ a large staff. According to collective agreements between public employers and trade unions, co-determination has been formalized through the establishment of local boards of cooperation public sector organizations. These bodies only have an advisory status. But in certain situations collective agreements directly stipulate that specific decisions are subject to negotiation between the management and local unions. In other situations, management has a strong incentive to make a deal with representatives of their own staff before making a decision. Such management behavior may be part of a strategy for guarding their own organization with regard to its institutional environment (Christensen Nannestad 1992). Public sector managers may have two reasons for adopting such a strategy. By taking care of the interests of their own staff, they cultivate support from them. In this way they may both add to their political strength vis-å-vis the environment and avoid conflicts with their own staff. Alternatively, experience may have taught them the wisdom of anticipating the reactions of their political principals. Hence, they realize that the formal right to make their own management decisions autonomously not protect them against critical reactions from politicians. Whether politicians directly intervene into their affairs is not important. Mere expression of political criticism of their management decisions is sufficient to make their job difficult. Either way interest organizations can be expected to be strongly involved in the implementation of policy.

To interest organizations, the most preferred situation is one where policy implementation is delegated directly to them. This is probably exceptional. But in sectors with strong traditions for corporatist decisionmakingit that administrative responsibilities are delegated to interest organizations. Such instances of delegated administration are found, for example, at the Danish labor market. They are also found within Danish agriculture and within the highly subsidized sector of cooperative housing. In these cases, interest organizations are typically given the responsibility for managing money transferred to them from the government. A similar

form of delegated bureaucracy is found in companies set up by local and central government, but regulated through corporate rather than public law. Important instances are public utilities, e.g. the natural gas companies established in the late 19705, and the government-owned companies with corporate status which have been made responsible for constructing bridges and tunnels across the Danish straits. In some cases, organized interests are represented on the boards of these hybrid companies. An additional attraction of delegated administration is its contribution to a depoliticization of administrative execution and to a reduction in the kind of transaction costs that might be a source of heavy overload (Heckathorn & Maser 1990).

Administrative Regimes and the Rule of Law

The above discussion of different forms of interest group involvement in public policy implementation provides a distinction between four distinct administrative regimes. These regimes differ from each other in important respects. Because of these fundamental differences, administrative staff responsible for policy implementation are also expected to behave in different ways. These differences are highlighted in Table 2.

Behavioral differences may be expected even though individual staff members are generally assumed to have preferences that are basically identical. Like other individuals, they are well aware of their self-interest. By intention they also act in a rational way. This rationality is clearly bounded; they operate, however, within the relatively stable framework of a specific organization, which again is rooted in a broader institutional setup of considerable stability. Because of this stability, which characterizes

important aspects of their professional life, they will normally be able to make up the balance in any course of action that might affect their own interests. For instance, in a professional context, these critical points may be related to their chances for promotion; they may also be related to their chances of having a peaceful workday that does not put them under undue stress.

Despite of this basic identifying of preferences that are assumed to motivate individuals, differences in an institutional context are expected to induce them to behave differently. This also applies to the four administrative developed above. First, these regimes will reward specific behavioral patterns differently. Variations in behavioral norms characteristic each administrative regime account for this. Second, their principles institutions of accountability will differ. Third, public exposure which contributes to opening up some administrative regimes may be entirely absent in other, more closed administrative regimes.

Pure and consultative bureaucracies share important characteristics in several respects. For one thing, both are governed by behavioral norms stressing public bureaucracy's respect of the principle of the rule of law. Although the institutions of consultative bureaucracy allow for affected interests to make themselves heard before making a final administrative decision, the primary obligation of public officials to apply the law when making decisions is not questioned. The involvement of interest group representatives in previous negotiations stresses their role as suppliers of information which has been defined as relevant arid legitimate by the law. Consequently, in consultative bureaucracy, administrative decisions are not presented as the outcome of a process of political bargaining. Of course, neither is this the case in pure bureaucracy.

Furthermore, in both pure and consultative bureaucracy individual staff members are subjected to combined judicial and parliamentary accountability. of how indirect the mechanisms of accountability may be, a comparatively high degree of public exposure creates an atmosphere in which it is difficult to hide malpractice and mismanagement. Consequently, individual staff members there is an increased risk that acts violating the norms which regulate their work will be revealed. Thus, other things being equal, it is reasonable to assume that individual civil servants responsible for implementing public policy will pay due respect to the law. This is the kind of behavior that is most likely to yield rewards valued by them. It is also a kind of behavior that does not involve any risks regarding their private preferences for keeping their job or just avoiding negative sanctions like criticism.

Corporatist bureaucracy, by comparison, involves organized interests
much more intensively in administrative decision-making than is the case
with consultative bureaucracy. Legal norms as laid down in a statute may still

be in place. But in some cases these legal norms are so broadly phrased that they do not give many prescriptions of any precision for policy implementation.This the case with some regulatory legislation, e.g. within the fields of environmental protection and workers' health and safety. More importantly,it to public policies where administrative implementation involves the spending of governmental funds. Such policies typically define a specific purpose, a set of criteria to be used as the basis of actual spending and distribution of funds, and a more or less well-defined limit on the amount of money involved. These policies may just distribute public funds among qualified recipients, be they private firms, private or semi-private, semi-publicorganizations even private households. They may also be the financial basis for the running of public and private organizations providing services for individual citizens and private firms. In virtually all such cases, the legal norms regulating policy implementation are vague, while the availability of public funds attracts intense attention from a set of directly affected and wellorganizedinterests. an institutional context like this, behavioral norms emphasizing mutual bargaining as the basis on which specific decisions are reached are quite likely to develop.

This institutional setup does not make the principles of judicial and parliamentary accountability inoperative. They will still apply. However, their effectiveness may suffer. The vague character of the legal prescriptions regulating administrative implementation will more or less empty the principles judicial accountability for specific content. This also happens with parliamentary accountability. As representatives of interest organizations and civil servants engage themselves in mutual bargaining as the basis for their administrative decisions, they also acquire a lot of autonomy vis-å-vis parliamentary institutions. Actually, this will be one of the attractions of corporatist bureaucracy for both group representatives and civil servants (Christensen & Christiansen 1992, 8-13).

Even more importantly, a new principle of accountability is introduced. One essential criterion for the evaluation of policy implementation under the administrative regime of corporatist bureaucracy will be its acceptance by affected interests. In this way, corporatist bureaucracy introduces a new political criterion for the evaluation of administrative behavior which replaces the principle of parliamentary accountability. Although legal provisionsfor accessibility of information may also apply in the case of corporatist bureaucracy, the prospects for keeping administrative decisionmakingclosed likely to increase. The administrative behavior expected under this administrative regime is that of accommodation. Public bureaucratswill strongly induced to oblige demands put forward by group spokesmen. They realize that normally they will not be criticized by their political masters for stretching themselves in this direction. The careermindedcivil may even have good reason to expect that such

accommodating behavior will be rewarded. Superiors who decide over
future promotions according to all experience will perceive it as an indicationof
valued qualifications.

With delegated administration, implementation of public policy is left to interest organizations or to other organizations placed outside the governmental Administrative decisions will still to some extent be regulated through externally fixed rules. But these rules will often be quite capacious. After all, the raison d'etre for delegated administration is to turn over a great deal of authority to organizations placed outside the governmental hierarchy. The implication is that implementation will be subject to intra-organizational norms. Under this regime, the leadership of the organization to which authority has been delegated will be responsible for policy implementation. In principle, they can be held accountable through judicial procedures. To some extent they may even be subjected to conventional governmental auditing. But once more it is to be expected that accountability will often be a purely intra-organizational matter. A conflict may in this case arise between externally set policy norms and intraorganizational The former will apply in cases of judicial and auditing accountability. The latter will apply if organizational leaders are held accountable for their behavior by their intra-organizational constituency. Such conflicts, however, are rare, as public exposure in these cases is low. The chances that possible irregularities could be revealed under these circumstances will be small. Therefore, it will be an exception that external procedures of accountability are activated. As to expected behavior, the result is that there is much room left for organization officers in charge of policy implementation to displace formal policy goals by other high priority goals defined by the organization. The incentive structures operating within these organizations interacting with the individual interests of these officers are expected to induce them to such behavior.

Administrative Regimes and the Mismanagement of Public Policy

Only pure bureaucracy, which fits into the notions of representative democracy,clearly with the relationship between general public policy and its implementation. As is well documented through implementation research and through research on street level bureaucracy, this in no way guarantees that the policy delivered is derived from the policy decided. Still, in these cases general policy will provide a set of comparatively uncontested standards for judging policy outcomes. In addition, there will be institutions of accountability of some effectiveness as there will also be

a rather high public exposure of the officers formally responsible for policy
implementation.

Of course, none of this precludes public bureaucrats pursuing other objectives than the ones set up through legal statutes. Such unauthorized behavior includes the introduction of self-generated routines and standards of the sort described by Lipsky (1980), budget- and slack-generating strategies in extreme cases, corrupt and fraudulent behavior. Pure bureaucracy, has to be seen as being placed on one extreme of a continuum. At the other extreme of this continuum is delegated administration. we move from pure bureaucracy towards delegated administration, norms are to an increasing extent substituted by other behavioral norms. It is also evident that public exposure is weakened. Institutions of accountability stressing some element of external control are superseded by principles of accountability which, instead, emphasize control as a matter of political responsiveness towards affected interests. Under these administrative regimes, individual staff members responsible for policy implementation will feel a much stronger temptation to complement even to replace official policy goals with other goals. Again, it is not corruption and fraudulent behavior in any traditional sense that are normally to be expected (Niskanen 1971, 192). Rather, it is opportunistic behavior where the combined effect of the rational pursuit of self-interest and opportunism is a mismanagement of public policy to the benefit of a well-organized caucus.¹

During the 1980s and the 19905, a number of such cases have been brought to public light in Denmark. In most of them the claim has been that public funds have been mismanaged. This has happened in two ways. One is that budgetary limits have been broken. By such overspending of funds, the restrictions contained in either the central government budget as voted by parliament or in the institutional rules constituting the central government budgetary system have been set aside. Another type of mismanagement the use of public funds for purposes that diverge from the purposes formally legitimated through the original appropriation.

There is plenty of information on these cases. Thorough investigations have been conducted, sometimes through several institutions. Public scandalshave been revealed through reports prepared by the National Department of Auditing. In some cases nothing more has happened. But one of the phenomena characteristic of Danish politics of this period has been a propensity to subject these cases to further investigation. The institutions used for these investigations have to some extent varied. But a recurrent solution has been to set up a court of investigation led by one or more judges. In these courts of investigation a wide range of civil servants, ministers, MPs as well as interest representatives presumably involved in the case are required to give evidence. After the investigation,

the judges have to prepare a public report setting out what, according to their judgment, happened and what might give grounds for legal criticism. The court of investigation is not asked to decide whether such criticism should end up in some form of legal prosecution or in political sanctions. Depending on the circumstances, this decision is left to parliament or to the minister responsible for the policy area in question.

Table 3 summarizes the scandals which, during the 1980s and the 19905, have (1) involved the management of public funds, and (2) have been subjected to public investigation. Several characteristics clearly stand out through the investigations which have been conducted. For example, only one of the scandals listed here took place within the institutional framework of public bureaucracy. Similarly, in only one case did the scandal take place within the institutional framework of a corporatist bureaucracy. In the remaining scandals, implementation was entrusted to organizations placed outside public bureaucracy. Consequently, in these cases the organizations to which implementation was delegated were not clearly integrated into the institutional hierarchy which theories of representative democracy perceive as essential to administrative and political accountability.

Pure Bureaucracy: The National Tax Agency

The case concerning unauthorized transfer of payments from the current budgetary year to the following year by the National Tax Agency seems rather trivial when compared to the other scandals. On the one hand, the breaking of budgetary rules applied in central government was unquestioned. the other hand, the case per se did not reveal overspending of the budget. By transferring the payment on the ADP-contract, the National Tax Agency imposed a financial burden on itself. In this perspective, the case sheds light on the stressful situation in which the management of the ministry was placed. Politically, the minister had strongly committed himself a grand reorganization of Danish tax administration. This, among other things, implied comprehensive budgetary and staff cuts to be implemented within a tight time schedule. Any deviation from this plan would add to the political criticism which had been directed towards the minister already when he launched it as his contribution to the governments campaign for "de-bureaucratization" (Statsministeriet 1988). Such deviations immediately be interpreted by the opposition as evidence of the minister's lack of control over his own organization and therefore from the perspective of the minister would constitute a political failure which he could hardly afford.

The director of the National Tax Agency seems to have clearly perceived
the high priority given to such considerations by the minister. To the court
of investigation, members of his staff explained that according to the

director's perception it was all-important to avoid any applications to the budgetary committee of parliament for supplementary appropriations (Beretning vedr. Skatteministeriet 1992, 882-883). In this context, they devised a solution which constituted a clear breach of budgetary rules. Their calculation may partly have been that this would expose them to less blame than a case of non-compliance with a rule, often perceived as an instance of bureaucratic formalism without substantive content. As they reported to the court, since 1984 the agency had occasionally made similar postponements of expenditures due to be paid. Eventually, this calculation turned out to be wrong. The principles of parliamentary accountability in this case were operative. The minister had to give up his cabinet post and disciplinary investigations were initiated against the director.²

Corporatist Bureaucracy: The Postal Scandal

By comparison, the postal scandal stands out as one of the biggest public sector scandals of this century. Problems accumulated over several years, dating back to the late 19605. Up until then, the Directorate General of Mail and Telecommunications stood out as an extremely centralized and very traditional public sector organization. During the 1960s and the early 19705, however, two developments occurred. First, the traditional civil service culture, which had characterized the postal services, gradually eroded. New types of staff were recruited. Strikes, which had previously been unknown (and which according to the civil service act were forbidden for civil servants), were suddenly used as a weapon for postmen. Absenteeism some places increased to an annual average of 30 percent or more. In Copenhagen in particular the postal organization went through radical change.

Second, a newly appointed top-management initiated a large-scale modernizationof organization. One goal was decentralization, another was the introduction of what was described as modern management methods, e.g. a system of management by objectives developed and implemented with assistance from a team of external management consultants hired at the Copenhagen Business School. However, both the traditional and the new postal service were permeated with corporatist norms. These norms laid down expectations that no issue of importance to postmen and other groups of staff would be settled unless negotiations had taken place between the unions, local shop stewards and the management. These negotiations were also expected to be concluded by an agreement if any change had to take place. These norms seem to have worked comparatively well during periods of stability. But with the changes in recruitment and staff structure, a radicalization of sectors of the employed staff coincided with an ambitious strategy for reorganization of the entire organization. The implementation

of this reorganization was explicitly based on a co-determination policy.
Although using an updated language, the co-determination policy was loyal
to the corporatist traditions of the postal services.

In order fully to understand the events that led to increasing management problems within the organization and to gross overspending as compared with the annual budgets voted by parliament, the political climate within which the organization operated must be taken into account. So, when the once acclaimed regularity of Danish mail services declined and the first strikes broke out,³ the message from the government was that resumption of work had top priority. In 1960 and again in 1968 the government intervened in the conflict. The way in which the intervention took place implied that it was seen as giving official government support to the unions (Beretning, post 1985, 230). In the latter instance the top management of the Directorate General experienced a Ministry of Finance that was more accommodating towards the unions than they had been themselves. Later, when the scandal was investigated, members of the management told the court about the impression this event had left on their minds. Together with civil servants from the Ministry of Finance they had, during the whole night, gone through a round of tough bargaining with the representatives of the unions. Next morning one of the civil servants from the Ministry of Finance had a personal call from the Prime Minister. He, the leader of a Liberal-Conservative government, only wanted to know whether the strike had been brought to an end. The economic implications of the deal made with the unions did nol: seem to concern him.

The message communicated by the Prime Minister was much like that of several ministers of public works who, over the years, were in charge of the postal services. "No minister wanted to see strikes within the civil service", a deputy director general reported to the court (Beretning, post 1985, 231). On later occasions, a similar political message was communicated the management. As a result, one of the sources for the soaring costs of the postal services was the introduction of performancerelated The system was based on a collective agreement with the unions. It was extremely favorable to the staff. Built-in mechanisms made it largely possible for postmen to decide their own pay. In this way the system made it difficult for management to control costs and thus to hold the budget. Still, when the question of renegotiation was opened, responsible ministers stressed the importance of not provoking an open conflict with the unions.

Delegated Administration: Private Administration of Public Funds

Delegation of administrative tasks to interest organizations constitutes a
further outgrowth of corporatist policy-making. To interest organizations,

the delegation of administrative responsibilities, such as the management and distribution of public funds, represents an ultimate victory. Through negotiations with their bureaucratic and political counterparts in public administration, they have obtained a precious concession. The minister in charge agrees to go to parliament with a bill delegating considerable poweis to them directly or to organizations affiliated with them, sometimes even directly controlled by them. Through this concession, the control of public funds is left to them (Meyer 1958, 211-213; Meyer 1964).

The cases concerning the agricultural funds and the Holiday Fund of the Labor Market are illustrative of this point. In these cases, a perfect interaction was established between public bureaucracy and private organizations. to relevant laws, a scheme was set up which actually introduced a tax-like levy on farmers and other private business dealing with agricultural products on the one hand, and entitled all wage-earners to receive holiday payments on the other. This provided the organizations involved with an independent financial source. Without backing from the tax-man, they would have been forced to collect the money themselves. Raising money through voluntary contributions even within their own constituency would have made them vulnerable to free-riding. The delegation authority to manage the funds thus received exempted them from considering all the procedures and other formalities of public bureaucracy. From their perspective, this is one of the attractive properties of delegated administration. Although both consultative and corporatist bureaucracy open public administration to interest group influence, group participation does not exempt civil servants from following basic bureaucratic procedures and other formal prescriptions characteristic of public administration. In the case of delegated administration, however, interest organizations are left to develop their own rules in a number of important respects.

In the situation which arose, private organizations could organize the management of the funds as they found appropriate. In the case of the agricultural funds, it led to a complex web of sub-delegation. Sometimes administrative responsibilities were transferred to agricultural interest organizations. In other instances, they were sub-delegated to organizations and institutions directly controlled by these organizations. The Holiday Fund developed a similar practice. On the average, 75 percent of its loans were given to the recreational organization Dansk Folkeferie, i.e. Holidays for the Danish People. The organization, which was set up as an independent was controlled by the Danish trade union movement and was placed at the center of a conglomerate with widespread business activities. The vertical integration of the conglomerate was strongly developed. controlled by the organization, therefore, provided all kinds of services for the holiday centers, and travel agencies were also run by the organization. This made the organization more or less independent of the market.

With delegated administration, goal displacement over time is a probable development. As the responsibility to manage public funds is delegated to private organizations legitimately pursuing other goals, its leadership is likely to divert funds from the original purpose to other goals pursued by the organization. A displacement of goals simply may take the form of profit-seeking behavior. The leadership of the organization perceives the opportunity to draw a profit from funds directly or indirectly placed in their custody. This opportunity was taken in both of the above cases. It was also taken in the case concerning the advance transfer of unemployment subsidies to the unemployment funds. The agricultural organizations could actually finance part of their interest group activities through financial support from the funds. Money, for example, was lent to the organizations without any demand for interest and repayments. This money was, in turn, invested in property controlled by the organizations. The same types of loans without obligations concerning interest and repayment were used in financial transactions between the Holiday Fund and Holidays for the Danish People. The unemployment funds similarly made a profit by placing the advance payments on accounts in which they were able to cash the interest themselves.

Lack of transparency, especially unclear rules of accountability regarding the ability to audit these organizations and their sub-contractors and beneficiaries, it difficult to reveal what happened. Owing to broad legal regulations, it was not clear where the limits were drawn between legitimate and illegitimate goals. This was even more difficult to uncover in these cases because strong cooperative bonds had developed between the organizations and relevant government departments and agencies. Neither the Ministry of Agriculture nor the Ministry of Labor found reason to interfere nor showed excess zeal when demands were voiced for pulling the strings a bit tighter around the organizations managing the funds.

Cases concerning the administration of public funds used to support housing for young people and financial subsidies to the youth organizations of the political parties share a number of basic traits with the agricultural and labor market funds. In these cases public funds were transferred to private organizations. These organizations, in other words, have also received authority to distribute funds to qualified recipients. Once more it is characteristic that the transfer of powers from public to private organizations not stop here. Tasks were further sub-contracted to other organizations affiliated with the organization, such as Ungbo, a Social- Democratic organization acting as an umbrella for a number of cooperative building societies providing housing to young people.

In one important respect, however, the latter two cases differ from the
cases concerning the agricultural labor market funds. In the former cases,
goal-diversion took place to the profit of the organizations to which delegationhad

egationhadoccurred. It was also questioned whether the financial managementof funds delegated to the organizations led to a financial loss for the government. Yet aside from these reservations, it was recognized that the organizations managing the funds transferred to them demonstrateda bureaucratic professionalism. Moreover, no claim was made in the cases of the agricultural and unemployment funds that the organizations had involved themselves in any criminal acts. Criticism was solely directed towards the branches of the civil service and the ministers in charge of the relevant policy areas. In the latter two cases, by contrast, the focus of criticism shifted to the organizations and their staff. In both cases police investigations were initiated to uncover whether individuals with management responsibilities in the organizations involved had committedcriminal liable to be punished.⁴ This also happened in the case concerning the holiday funds. Only here, allegations were made that individual managers were illegally profiting from transactions financed through government funds. But upon investigation, the public prosecutor found no grounds for bringing the case to court.

Conclusions

The four types of interest group involvement in public policy implementation different administrative regimes. The basic difference between them concerns the weight they place upon the rule of law as opposed to the rule of affected interests. In his classical study of liberal democracy, Lowi pointed to the risk that extended involvement of interest organizations might endanger essential values in democracies which strongly emphasize the rule of law (Lowi 1979). However, the debate opened by Lowi focused primarily on American politics and administration (see e.g. Schaefer 1988; Lowi 1988).

For European countries, the debate over the role of interest organizations in democratic society has taken quite another direction. Neo-corporatism came to dominate most theoretical discussions from the 1970s and onwards. So, critical questions from the very beginning of the neo-corporatist wave were raised as to its descriptive accuracy and its lack of coherent theoretical reasoning (Heisler 1979). Other writers focused on its presumed negative or positive consequences for economic growth (Olson 1982; Katzenstein 1985). Quite another group of writers stressed corporatism's positive contribution social and political stability. Their claim is that this is the standard by which corporatist structures have to be measured (Lewin 1992).

However, neo-corporatist and corporatist contributions suffer from
severe weaknesses. It is never quite clear what the theoretical reasoning
behind the corporatist model is. So, it is not possible to find out why one

should expect this particular behavior instead of another (Nannestad 1991, 33-36). Furthermore, the neo-corporatist theory suffers from the typical weaknesses of macro-theory. Its strength is measured by its ability to set the agenda for a broad debate among political scientists. Its weakness, however, becomes clear when it comes to the specification of the relevant variables and their operationalization. Similarly, the weakness of corporatisttheory out in its negligence of any discussion of the mechanismswhich connect the ill-defined independent variables intended to catch important aspects of modern industrialized societies with policy implications of specific content and direction. In this context, it is telling that one of the recent contributions to the debate over corporatism fails to specify the empirical scope of corporatist structures as defined by Schmitter in the early 1970s (Schmitter 1974, 93-94; Lewin 1992). Is corporatism in this sense a distinct feature of a highly organized labor market, of the interaction between private industry, the unions and government with specific relevance for the study of industrial policy? Or is it a generalized description of the institutionalized interaction between any kind of economicinterest and any branch of government? A consistent answer is not given.

In this article, corporatism has been defined in a less restrictive way. Corporatism is seen as the integration of organized interests in the making and implementation of public policy. This leaves open the forms which corporatism may take. It even leaves open whether the ideal-type definition of neo-corporatism has any empirical validity. Whatever its specific forms, separate problems concern corporatism when it is treated as the dependent variable and when it is treated as the independent variable. In this article, corporatism has been dealt with as the independent variable. At a general level, the article has pursued a double goal. First, it has been shown that to understand the role played by interest organizations in Western Europe, it is necessary to deal much more explicitly with governmental structures and forms of organization. Surprisingly., when the original conceptualizations corporatism are considered, this has generally been neglected by most writers on corporatism (cf. e.g. Meyer 1958 and 1964; Offe 1981; Schmitter 1985). When eventually state and government have been considered in the writings of neo-corporatist political scientists, these institutions been subjected to macro-level analysis, which makes it difficult to draw any testable conclusions. Second, it has been demonstrated that corporatist structures have consequences for policy implementation.

More specifically, the analysis has focused on a number of public scandals which have been uncovered during the 1980s and the early 1990s in Denmark. In all these cases, criticism was raised against the mismanagementof funds. Corruption was not involved in any of the cases. No instances were revealed where individual civil servants received

any monetary rewards or other specific forms of appreciation from clients as the payment for making decisions favourable to their interests. The analysis has instead addressed a much more subtle phenomenon. This concerns the institutional setup of different administrative regimes. They differ from each other with respect to the form and intensity of interest organization involvement in administrative implementation of public policy. Here, the study found support for several conclusions:

(1) With increased involvement of interest organizations in the implementation public policy, legal norms are superseded by political norms in as far as concern the guidance and the legitimization of administrative behavior.

(2) As interest organizations are increasingly involved in policy implementation,
phrasing of public policy decreases in precision as to the
prescriptions given to the administrators in charge of its execution.

(3) With increased involvement of interest organizations in the implementation public policy, judicial and parliamentary institutions of accountability characteristic of representative democracy are weakened.

(4) The weakening of parliamentary and judicial institutions of accountability a consequence partly of the lack of precision in policy prescriptions, of an increased political reluctance to hold the implementing staff accountable for their administrative decisions.

The case studies included in this analysis draw attention to delegated administration. As a matter of fact, most of the public scandals that have hit Danish central government over the past decade took place within an administrative regime of delegated administration. According to the propositions developed in this article, this is in no way surprising. Although an interesting feature of a democratic system with strong corporatist traits, delegated administration empirically constitutes an exception. In empirical terms, the administrative regime which was defined above as corporatist bureaucracy is much more interesting. This applies both to the precise forms in which it operates and to its effects.

It seems appropriate, therefore, that further studies of interest group involvement in public policy implementation should focus on corporatist bureaucracy as it has developed within the welfare service sector. This sector makes up a large part of the public sector in Denmark as well as in the other Nordic countries. Institutions of this sector, moreover, constitute the dominant basis of government employment. To a very large extent the employees of the welfare sector are unionized, which places the unions in a strong political situation. Finally, inasmuch as these welfare institutions are typically run by local and regional governments, research in the influenceof

enceofinterest organizations on the management of public funds should be directed towards this level of government. This research would involve a general analysis of the role of unions in the management of public sector institutions. The interaction of the unions with local managers in this context, together with in-depth studies of the institutional setup within which public sector organizations operate, would be of particular interest. The expectation is that one would find an institutional system within which managers feel strong incentives to disregard considerations of costs and efficiency.

---

NOTES

An earlier version of this article was presented at the workshop on changing forms of corruption, at the ECPR Joint Sessions of Workshops, Leiden, April 1993.

1. Opportunism is here used as defined by Williamson (1985, 47-48): "By opportunism I mean self-interest seeking with guile. This includes but is scarcely limited to more blatant forms, such as lying, stealing, and cheating. Opportunism more often involves subtle forms of deceit . . . More generally, opportunism refers to the incomplete or distorted disclosure of information, especially to calculated efforts to mislead, distort, disguise, obfuscate, or otherwise confuse. It is responsible for real or contrived conditions information asymmetry, which vastly complicate problems of economic organization."

2. The director, incidentally, lost his job because of problems connected with the implementation of the reorganization plan. The reorganization which involved the merger of two agencies led to intense conflicts within the new organization. These conflicts, which also took place at the management level, influenced the climate within which the scandal developed.

3. Actually, they were the first strikes among public employees with civil service status.

4. The criminal case concerning the fabrication of false enrollment registers in the youth organizations of the political parties is still running.

REFERENCES

Christensen, B. 1958. Råd og nævn. Copenhagen: Gad.

Christensen, J. G. 1980. Centraladministrationen: Organisation og politisk placering. Copenhagen:
Forlag.

Christensen, J. G. 1991. Den usynlige stat. Copenhagen: Gyldendal.

Christensen, J. G. & Christiansen, P. M. 1992. Forvaltning of omgivelser. Herning: Systime.

Christensen, J. G. & Nannestad, P. 1992. The Institutions of the Welfare State. Aarhus:
Institute of Political Science.

Day, P. & Klein, R. 1987. Accountabilities. London: Tavistock Publications.

Heckathorn, D. D. & Maser, S. M. 1990. "The Contractual Architecture of Public Policy: A
Critical Reconstruction of Lowi's Typology", Journal of Politics 52, 1101-1123.

Heisler, M. O. 1979. "Corporatist Pluralism: Where Is the Theory?", Scandinavian Political
Studies 2, 277-297.

Katzenstein, P. J. 1985. Small States in World Markets: Industrial Policy in World Markets.
Ithaca: Cornell University Press.

Kristensen, O. P. 1979. "Centraladministrationen og de særlige forvaltningsorganer i Danmark
efter 1945", Nordisk Administrativt Tidsskrift 59, 69-98.

Lewin, L. 1992. Samhållet och de organiserade intressena. Stockholm: Norstedts.

Countries: Conceptual and Empirical Linkages", British Journal of Political Science 21,
235-256.

Lipsky, M. 1980. Street-Level Bureaucracy. New York: Russell Sage Foundation.

Lowi, T. J. 1979. The End of Liberalism. The Second Republic of the United States, 2nd edn.
New York: Norton & Co.

Lowi, T. J. 1988. "Response to Schaefer", Administration and Society 19, 399-412.

Meyer, P. 1958. Politik. Statskundskab i grundtræk. Copenhagen: Nyt Nordisk Forlag Arnold
Busck.

Meyer, P. 1964. "Organisationernes deltagelse i forvaltningen". Nordisk Administrativt
Tidsskrift 45, 381-387.

Moe.T. M. 1989. "The Politics of Structural Choice", in Chubb, J.E. & Peterson. P.E.,eds.,
Can the Government Govern? Washington: Brookings.

Nannestad, P. 1991. Danish Design or British Disease. Danish Economic Crisis Policy 1974-1979
Comparative Perspective. Aarhus: Aarhus University Press.

Niskanen. W. A. 1971. Bureaucracy and Representative Government. Chicago: Aldine-
Atherton.

Offc, C. 1981. "The Attribution of Public Status to Interest Groups: Observations on the
West German Case", in Berger, S., ed., Organizing Interests in Western Europe. Cambridge:
Cambridge University Press.

Olson, M. J. 1982. The Rise and Decline of Nations. Economic Growth, Stagflation, and Social
Rigidities. New Haven: Yale University Press.

Schaefer, D. L. 1988. "Theodore J. Lowi and the Administrative State", Administration &
Society 19, 371-398.

Schmitter, P. C. 1974. "Still the Century of Corporatism?", in Pike, F.B. & Stritch, T., eds..
The New Corporatism. Social-Political Structures in the Iberian World. Notre Dame:
University of Notre Dame Press.

Schmitter. P. C. 1985. "Neo-Corporatism and the State", in Grant, W., ed., The Political
Economy of Corporatism. London: Macmillan.

Statsministeriet 1988. Handlingsplan for afbureaukratisering. Copenhagen: Prime Minister's
Office.

Weber. M. 1922. Wirtschaft und Gesellschaft. Tubingen: J. C. B. Mohr. 5. rev. ed. 1972.
Williamson, O. E. 1985. The Economic Institutions of Capitalism. New York: Free Press.

Documentation of the Cases Discussed
Tax administration

Beretning afgivet af komissionsdomstolen vedr. visse forhold inden for Skatteministeriets
område. Copenhagen: 1992. Vols. 1-2.

Fostal scandal

Beretning over den af statsrevisorerne afgivne beratning af 6. august 1981 om post- og
telegrafvæsenets ressourceforbrug og ressourcestyring i 1980. Beretning afgivet af finansudvalget
4. september 1981.

Beretning om undersøgelse af visse forhold vedr. post- og telegrafvæsenet. Afgivet af den i hh.
t. lov nr. 85 af 17. marts i982 nedsatte kommissionsdomstol. Copenhagen: 1985, Vols. 1-

Agricultural funds

De af Folketinget valgte Statsrevisorer. Beretning nr. 23 1983 om produktions- og promilleafgiftsfondene
for landbrugs-, gartneri- og fiskeriområdet.

-. - nr. 24 1983 om visse finansieringsfonde m.v. inden for landbruget.

- nr. 17 1988 om smittebeskyttende foranstaltninger ved landegrænsen og programmet til
bekæmpelse af kvægsygdommen IBR.

-. - nr. 2 1989 om finansierings-, afgifts- og sikringsfonde for landbrug, gartneri og fiskeri.
Beretning afgivet af Kommissionsdomstolen vedr. Landbrugsfondene. Copenhagen 1991.

Unemployment funds

De af Folketinget valgte statsrevisorer. Beretning nr. 18 1979 om udbetaling og forrentning af

statens forskud til arbejdsløshedskasserne. Copenhagen 1980.

Redegørelse om statens forskudsvise udbetalinger til arbejdsløshedskasserne til dækning af
udgifterne til arbejdsløshedspenge m.v. Vols. 1-4. Copenhagen 1984.

Beretning om undersøgelse af statens forskudsvise udbetalinger til arbejdsløshedskasserne til
dækning af udgifterne til arbejdsløshedsdagpenge m.v. Copenhagen 1985.

The Holiday Fund

De af Folketinget valgte statsrevisorer. Beretning nr. 3/90 om Arbejdsmarkedets Feriefonds

støtte til Dansk folke-ferie til anlæg af feriecentre mv. Copenhagen 1992

The Ungbo scandal

De af Folketinget valgte statsrevisorer. Beretning nr. 6/91 om forvaltnignen af og tilsynet med

statsstøtte til ungdomsboliger. Copenhagen 1992.

Political youth organizations

De af Folketinget valgte statsrevisorer. Beretning nr. 1/91 om tilskud fra tipsmidlerne udbetalt

gennem Dansk Ungdoms Fællesråd til politiske ungdomsorganisationer. Copenhagen 1992.