Resumé

There has been a growing concern about administrative reform in many countries during the las: 10-20 years. The central administrative apparatus has been restructured, partly as an instrument to fulfill collective, political goals. This study of the reorganization of the central health administration inNorway shows that political and administrative leaders increase their control through a planned reorganization. But such a comprehensive reform also creates problems in adjusting to new administrative roles.

Introduction

As part of a 'public revolution' the organization of the administrative apparatus in many countries has changed substantially after World War II (Olsen 1986). Public administration is now more multi-structured and nonhierarchical than before, and public authority is more decentralized.

Governments increasingly produce reports and proposals on reorganization, and administrative bodies with special responsibility for administrative reform have been established (Egeberg 1984; Pollitt 1984). The institutionalization of structural development has secured this relatively new field a stable administrative attention and access to the public agenda (Egeberg 1985).

Studies of administrative reform can be divided into two parts: Analyses of reorganization processes, including comprehensive administrative reform (March & Olsen 1983; Pollitt 1984; Seidman 1980) as well as case studies (Grafton 1979; Garnett & Levine 1980; Roness 1982; Egeberg 1984; Christensen 1985), and analyses of reorganizational effects (Egeberg 1984; Christensen 1985). Methodological and empirical problems as well as little concern with evaluation and learning in administration may explain the limited number of effect studies (Salamon 1981). The latter reason may be due to capacity problems, trained incapacity, internal conflicts, fear of failure and symbolic politics (March & Olsen 1983).

One central question in the reorganization studies is to what extent, and
how, political and administrative elites use administrative reform to further
political control and collectively approved goals.

This article refers to this problem by focussing on the comprehensive

---

*This article is based on Christensen (1985).

reorganization of the central health administration in Norway. The

main

questions to be answered are the following: How did this restructuring come about? Did the political and administrative elite control the process by executing hierarchical authority in an instrumental way? What are the consequences of the reorganization for the individual and organizational decision-making behaviour in the health administration? What is the connection between intentions and effects?

Perspectives on Administrative Reorganization

Most studies of reorganization processes in central administrative institutions agree on one empirical fact: Reorganization is initiated, controlled and enforced by political and administrative elites (Roness 1982; March & Olsen 1983; Christensen 1983; Egeberg 1984; Pollitt 1984).¹

There are different ways to interpret this fact, however, leading to different
perspectives on reorganization.

From one theoretical perspective the reorganization of the central administrative apparatus is, primarily, an instrumental activity. There are collectively defined goals for public institutions and some of these can be fulfilled through a restructuring (Egeberg 1985). The organizational thinking is of a typical means-end character. It is supposed that new administrative guidelines or constraints alter people's beliefs and their administrative behaviour. Thereby administrative bodies modify their activities, and the content of public policy in different fields changes.

According to this instrumental perspective, reorganization processes are planned and controlled. Political and administrative leaders have, through hierarchical positions, both the responsibility for administrative reform and the means to control these processes. The decision-making is characterized by exclusive participation and clear-cut procedures.

The instrumental perspective on reorganization is a rather common one. Olsen (1986) formulates 'the sovereign, rationality-bound state' model on administrative reform, stressing the political leaders as rational 'designers' furthering collectively approved and a priori defined goals. March & Olsen (1983,282) call it the rhetoric of orthodox administrative theory:

This rhetoric speaks of the design of administrative structures and procedures to facilitate the efficiency and effectiveness of bureaucratic hierarchies...The rhetoric of administration proclaims that explicit, comprehensive planning of administrative structures is possible and necessary, that piecemal change creates chaos.

The same theoretical thinking is represented in the 'machinery of government' metaphor (Hood 1979; Eriksen 1986) and in Pollitt's analysis of the importance of 'organizational designers' in administrative reform in England (Pollitt 1984).

From another theoretical perspective, a conflict-coalition perspective, reorganizing public administration is decided upon through struggle, conflict, consultations and negotiations among actors inside and outside the public apparatus (March & Olsen 1983). The actors in such collegial structures have different goals and interests, and different views on administrative restructuring.

Research has shown that public administration is rather heterogeneous regarding social biography, bureaucratic careers, bureaucratic milieu, opinions, contacts, etc. (Lægreid & Olsen 1978; Olsen 1983). Administrative units compete and political control is more difficult to obtain than previously.

In many countries a democratization of public policy and administration has taken place. A central aspect of this trend is that integrated participation in government by interest groups has gained in strength and formalization (Olsen 1981). Civil servant organizations have acquired rights of participation in administrative decision-making, including decisions on structural reform.

According to this perspective, binding decisions on administrative reform are reached through compromise or winning coalitions. A new structural plan reached through a negotiation process may be characterized by the same means-end thinking and organizational solutions as a plan reached through hierarchical control. The former process may generally be more timeconsuming, but may also give a firmer basis for implementation. On the other hand, reorganization processes characterized by conflict and negotiation may lead to ambiguous reorganization plans not very well suited to fulfill official goals and change administrative behavior.

Our analysis of reorganization effects is based on the instrumental perspective. According to this perspective, the rationality in public administration is founded on the total organizing of units and roles, not related to the single decision-maker (Scott 1981). The behaviour of all actors taken together and shaped by the structure further the goals of the organization (Egeberg 1984).

Public institutions can consciously, on the basis of clearly stated goals, be designed and redesigned. At a certain time organizational units, actors, issue areas and tasks can be coupled to obtain certain effects, but at another consciously be kept apart.

The organization structure in public administration decides duties and rights concerning tasks and decisions, actors, definition of problems, administrative technology, and acceptable solutions. Certain mechanisms secure that formal structure really determines behaviour (Lægreid & Olsen 1978). One is that actors are socialized into the administrative milieu. Another is to discipline actors by promising incentives such as lifelong careers and gradually escalating wages and prestige. A third mechanism is direct control and programming of people in administration by the leaders.

If the structural lines of guidance or constaints are altered the result will be
a change in the decisions of the individual decision-makers and eventually in
the decisions of the administrative apparatus.

The instrumental perspective tries to isolate the effects of changes in the organizational structure - 'the manipulation variable' - in comparison to other independent variables affecting decision-making behaviour and the content of public policy (Egeberg 1984). Example of other possible independent variables are physical structure, organization culture, organization demography and organizational environment (Egeberg 1985; Davis 1984; Goodsell 1977; Pfeffer 1982).

The instrumental perspective is based on constitutional theory and classical theories of administration (Weber 1970; Gulick 1969). Comprehensive empirical studies of public administration in recent years have shown that there is a strong relationship between structure and behaviour (Mayntz & Scharpf 1975; Lægreid & Olsen 1978; Grønnegaard Christensen 1982; Olsen 1983). Lægrid & Olsen (1978) show, for example, a strong positive connection between formal structure and patterns of contact in Norwegian ministries. People in the ministries have far more contact inside than outside their own sector and ministry. The pattern of contact varies with hierarchic level, i.e. leaders have systematically the highest scores on all measures of contact. In the larger ministries the frequency of contact between top and bottom is lower than in smaller ones.

But one thing is to show a strong connection between structure and behaviour, another to prove that structure really shapes behaviour. Egeberg (1984) shows in an analysis of reorganizations in Norwegian ministries in the 70s that the decision-makers' contacts change when the structural constraints are altered. The changes in contacts are largely those predicted within an instrumental perspective. Egeberg also found that reorganizations resulted in new or changed goals, new perspectives or new tasks for a lot of people in the ministries. He also demonstrated that the merging of organizational units created more ambiguity and conflict than the dividing of them.

An Empirical Review

The development of the central health administration in Norway is characterized
by a variety of organizational forms, of which the independent
directorate has been the most common.2

Historically speaking, the political and administrative leaders (from the legal profession) preferred control through administrative structures inside the ministry. But the medical profession stressed professional autonomy and administrative freedom outside the ministry. The penetration of the independent

health directorate in 1875 shows the strong position of the medical
profession in reorganization processes.

After World War II the government changed its administrative doctrines. It now preferred to establish more independent directorates. This was mainly an answer to capacity problems. By hiving off the directorates, political leaders aimed to increase their control through small and politically strong ministries.

At the same time the independent Health Directorate was reorganized, but in another direction. In 1945 it was given a semi-autonomous position: The Director General should partly be a leader of an independent directorate, partly a senior government official in the ministry heading a department. This more and more uncommon administrative construct was retained until the reorganization in 1983 (Fig. 1).

The medical profession hoped at the same time to maximize political access and professional autonomy with this new structure. It implied direct access by the Director General to the minister and a pushing back of the legal profession from central positions in the health administration.

The semi-autonomous position of the directorate, combined with an unchanged internal structure, a substantial growth and an earmarking of leadership positions for medical doctors, gave in fact the medical profession a unique opportunity to define and direct the development of the national health service for nearly 35 years.

During the 70s the need for a reorganization of the directorate became more and more obvious for many reasons. A new law had placed the responsibility for hospitals at the county level and a decentralization of the primary health care was under debate. Economic stagnation led to demands for a more effective health sector, including a closer coordination between political, economic and professional values.

Internal problems in the directorate included capacity problems, fragmentation,
and a growing dissatisfaction with a career-system blocked by the
medical profession.

In 1981 the Labour government in Norway was replaced by a Conservative minority government. In the spring of 1982 the government decided to ask the Under Secretary of State in the Ministry of Health and Social Affairs to work out a new organizational structure for the ministry, paying special attention to the health administration. Her report was based on an organizational model proposed in 1981 by the administrative leaders in the ministry. Her proposal was met by severe criticism from the Health Directorate, but was well received by other ministries. In October 1982 the government decided to adopt the proposal as its own policy. The directorate should be divided: The smaller part was to be established inside the ministry as a health department, the larger part to be moved out of the ministry and established as an independent health directorate. The proposed internal reorganization of the directorate was comprehensive too. The former divisions were to be replaced by departments and sections; persons, positions and tasks would be reshuffled.

The period between autumn 1982 and autumn 1983 was an elaboration and implementation phase. The process was opened up; several participants with different views were invited to give their opinions on the elaboration of the new organizational structure. But the frames were fixed by the governments' decision, and there were only minor issues left to be settled.

The political-administrative leadership fixed a certain deadline for the
implementation of the new structure, leaving no opportunity for an obstruction
by its opponents in the directorate.

The Reorganization Process: Political and Administrative Control

We have described a reorganization process controlled by political and administrative leaders in the Ministry of Health and Social Affairs. This control concerned both the initiative, the decision, the elaboration and the implementation of a new organization structure.

From what perspective is this controlled reorganization process

best

understood? The first question to be asked is this: How could the political
leadership take control over the first phases of this process?

From an instrumental perspective the answer is quite simple: The political leaders in the ministry wanted to do something with the structure of the central health administration. During the 70s there had been a growing dissatisfaction with the directorate and its leader among the non-socialist parties. They wanted more political and economic control in the health sector and thought the Director General was more of a politician than a senior government official.

The political leaders decided to allocate time, attention and resources for reorganizing the directorate. The Under Secretary of State was heavily engaged in the process: a typical political priority. The process had very tight deadlines, thereby securing political control. The politicians in the ministry also knew that their initiative would meet little resistance in parliament. The Labour Party seemingly wanted a reorganization.

A special condition was the fact that the political leaders had professional
competence in the field. The minister was a professor of odontology, the
Under Secretary of State a professor of medicine.

In sum, the political leaders in the ministry had both political will, legitimacy and professional competence, all in accordance with the instrumental perspective on reorganization. But they had no clear idea of which organizational model to choose. And they had not secured firm support for the implementation of a new structure. In this situation they had to find coalition partners, as one could predict from a coalition-conflict-perspective.

It was a quite understandable choice to make an alliance with the administrators in the ministry. The administrative elite had long wanted a reorganization and previously proposed the model now adopted by the political leaders. The administrators had the administrative means to back up the politicians, and soon chose to allocate resources to deciding upon, detailing, and implementing a new structure.

The 'ministry-coalition' totally dominated the first phases of the reorganization
process. But how could this happen against the interest of the medical
profession and the civil servants' organizations in the directorate?

In a way the ministry-coalition turned the decision-making process upside down. First, the conclusions were reached, then the premisses worked out. A normal procedure would have been to set up a representative commission to discuss a reorganization of the Health Directorate. Instead, the ministrycoalition decided to close the process, leaving the opponents on the sideline.

The opponents were mainly gathered in a 'directorate-coalition', a more
loosely coupled unit than the ministry-coalition. The medical profession in
the directorate, supported by most doctors in adminstrative positions outside

the

the directorate, protested against both the procedure used and the decision reached. They wanted to maintain the existing organization structure with some minor internal changes in the directorate. The employee organizations in the directorate wanted to participate in the process right from the beginning, but they were not totally negative to the structure proposed.

The Under Secretary of State refuted the demands from the directoratecoalition. She said that the reorganization had to be defined as a major change in the structure of the ministry, and that it was a question to be decided by the political leadership. This was obviously a legitimate view, but certainly not the only one. She promised that the different interests involved could take part in the next phase of the process, the elaboration of the new structure chosen by the government.

The opening up of the process in the elaboration phase can be interpreted as a strategic manoeuvre from the winning coalition to secure control and legitimacy through following the rules of the game and creating a feeling of democracy. They also hoped that the participation from the opponents created a better understanding of why this kind of reorganization was necessary.

The directorate-coalition participated in this phase with mixed feelings. From one point of view, the opening up of the process was democratic decor, but from another there was always a hope of obstructing previous decisions. The employee organizations had been struggling for the right to participate in inner decision-making processes and wished to use this right, even if their influence would be minimal.

Given the high conflict potential, two kinds of problems would have to be faced in the implementation phase: First, the opening up of the process, without substantial changes in the reorganization plans, could provoke the opponents and strengthen their criticism of the procedures. Second, the opponents could muster new arguments on cursory organizational thinking. All this together could be a major obstacle to the adaption and proper working of the new structure.

In sum, the ministry-coalition emphasized action and control more than organizational thinking in the process. Had the objections from the opponents been taken more seriously early in the decision-making process, there probably would have been no independent directorate and more time and energy spent in negotiations.

Effects of the Reorganization

From the instrumental perspective we can formulate three assumptions
concerning the effects of the reorganization on the decision-making behaviour
of actors in the central health administration:

- there will be a strong, positive relationship between the structural
changes and the individual decision-making consequences.

- as a whole there will be greater individual decision-making effects on actors in the health directorate than in the health department. This is because people in the directorate experience the most profound formal changes, i.e. a combination of changing the authority lines to the ministry, diving the former directorate and an inner reorganization of the directorate ³

- there will be variations in consequences within both of the two organization units concerned, because of the fact that the actors are differently affected by the formal reorganization. Some of them experience that their administrative lives are turned upside down, some are placed in almost the same administrative milieu before and after the reorganization.

The analysis of the effects of the reorganization is based on the major structural changes that were made in the central health administration. We use opinions on changes in the actors' contacts and working conditions as indicators of reorganization effects (Laegreid & Olsen 1978; Egeberg 1984).⁴

The main question to be answered is then reformulated: To what extent
and how were the actors patterns of contact and working conditions affected
by the major structural changes in the reorganization?

First, what kind of consequences for the contacts of the individual decisionmakers concerned should one predict from dividing the former directorate and moving the new one out of the ministry? One expectation, according to the instrumental perspective, is that people in the directorate moved out of the ministry will lose some contact with the political leaders, other departments in the ministry, and other ministries. The greatest consequences, are expected in the most demanding contact, that is with the political leaders. In contrast, decision-makers who, through the reorganization, move from the directorate to the health department are expected to strengthen their contacts on all these variables.

Table 1 shows a close relationship between the formal structural changes
and the consequences for the pattern of contacts.

According to the expectations derived from the instrumental perspective, the reorganization has profound consequences for the patterns of contact of the individual decision-makers. Employees in the directorate, as a whole mostly affected by formal structural changes, experience a major weakening of their contacts as a result of the reorganization. Taken together the contact is weakened on at least one of the measures for 69% of the respondents in the directorate. The most important effect is that many lose their contacts with the political leaders in the ministry.

The people who move from the directorate to the health department are

formally closer to the political leaders and other organizational units than before, but the change is not dramatic. This is reflected in the changes in the contacts they experience. Primarily, some of them strengthen their contacts, as one would expect. The most important is that they gain in political importance what people in the directorate lose through the reorganization.

The changes in the pattern of contacts after the reorganization take place within the same physical structure, i.e. the employees in the two organizational units are situated in the same building as before the reorganization. This underlines that the new formal structure to a large extent reshapes the contacts of the people involved.

The effects on the pattern of contacts may show one important condition for strengthening the political coordination in the health sector. The reorganization aimed at a hiving off of the directorate, thereby giving the political leaders in cooperation with the health department more power compared to the professional groups in the directorate, and this seems to have been fulfilled. Our data indicate a growing political isolation of the directorate according to the intentions behind the reorganization. The directorate takes on a more typical professional role, i.e. it does not have to take political or economic conditions into account. This is also a change from a role of aggregation to one of articulation.

Another major component in the reorganization was the comprehensive
internal restructuring of the directorate. This underlines that people in the
directorate had to cope with many formal changes at the same time.

One of the internal changes concerning the pattern of contacts was the establishment of a new hierachic level in the directorate. A substantial number of the respondents in the directorate report that this change resulted in fewer contacts with the leadership in the directorate. This is in contrast to the respondents in the health department who experienced a closer relationship to their leaders within a much smaller organizational unit.

Taken together, the four different measures of contacts show a typical
cumulative pattern. The respondents in the directorate are heavily overrepresented

in the group reporting a weakening of contacts on the majority of the measures. And people from the health department dominate in the group of answers emphasizing a strengthening of contacts. This polarization in consequences, due to a major difference in type and strength of formal structural changes, is in accordance with predictions from the instrumental perspective.

But this does not mean that all the employees in the two units wen; directly affected by the reorganization. Some were almost untouched by the reorganization because they kept the same positions and tasks. These respondents therefore report few experienced effects. This strengthens the explanatory power of the instrumental perspective.

Since our data concern the opinions on the effects of reorganization, we shall control for formal changes.⁶ Do our data show the 'real' consequences of reorganization, or do they just indicate personal opinions or events in the reorganization process? The answer to this is quite simple. There is a strong positive relationship between the degree of formal changes and the effects reported. Many of the respondents in the directorate formally lose their former administrative milieu through a hiving off and an internal reorganization, and they also report the most dramatic changes in their contacts with the political leaders in the ministry, other departments in the ministry, other ministries and with their own leaders.

Our next question concerns the effect of reorganization on the administrative milieu of the individual decision-maker. Did the reorganization lead to new or changed goals, tasks, priorities and perspectives? Did the employees experience a substantial change in the expertise represented? Did the reorganization result in more clear-cut goals and a lower level of conflict than before?

Generally speaking one would predict that these kinds of effects should be
substantial. The formal structural changes were comprehensive, especially for
those who remained in the directorate.

The people involved in the reorganization experienced major changes in their work. 88% of the respondents from the directorate and 78% from the health department report that their administrative milieu has changed one way or another. Table 2 shows that the changes in goals and tasks are substantial in the two new units, reflecting that the employees have had to adjust to new roles: the people in the directorate to a more professional one, the people in the department to a more political one. Respondents from the health department score higher on new expertise, which is due to the fact that this is a brand new unit composed of element from different divisions in the former directorate.

It is obvious from Table 2 that the respondents in the directorate
experience negative consequences from the reorganization. More than half of

the respondents say that there are less clear-cut goals and more conflicts after the reorganization. This is due to many factors. One is ambiguity in the allocation of authority between the directorate and the health department. A second is the lack of influence of the directorate on the decisions taken in the department. A third is the ambiguity connected with the effects of the new internal structure in the directorate. This illustrates that dividing an organizational unit may be evaluated in negative terms when coupled to other problematic structural changes. Normally dividing is more welcomed than merging by the actors involved (Egeberg 1984).

As for the contacts, the effects of the reorganizations on the constraints of the decision-makers show a cumulative pattern. A substantial part of the respondents from the directorate experience many changes at the same time. As a whole these are negatively evaluated. The respondents from the department are more divided in their opinions.

The dramatic changes experienced by the decision-makers in the directorate
are mainly in line with the predictions made from the instrumental
perspective. But they can be interpreted in different ways.

From one point of view they reflect the intentions behind the reorganization; a strengthening of the central coordination and political control. Administrative milieus were broken up and new ones established to fulfill these goals. The employees were confronted with new decisional constraints. In this way the reactions reflect a time of transition before the new structure is more established.

Another point of view is that the reorganization has been too comprehensive; it has changed too much for the individual decision-maker. The result is ambiguity, frustration and conflicts. Many of the people in the directorate have problems in adjusting to the new structure. The aggregated results of this can be other effects than those intended.

Summing up, the two main elements in the analysis show the pattern
predicted by the instrumental model. A substantial part of the respondents
from the directorate face different formal changes through the reorganization

and report that their range of contacts has been weakened and their working
conditions changed in a negative way.7

The respondents in the department face less substantial formal changes and their experiences with the new structure are less clear-cut. They tend, however, to be satisfied with the new structure because they are now placed in a smaller unit with greater political influence, more sharply defined goals, less conflicts, and more challenging tasks.

Conclusions

This article has focussed on the comprehensive reorganization of the central
health administration in Norway in 1983, with respect to both the decision -
making process and the effects.

The reorganization process is analyzed by combining an instrumental and a conflict-coalition perspective. A coalition of political and administrative leaders in the Ministry of Health and Social Affairs controlled the reorganization process. Representatives from the Health Directorate, the institution most directly involved, had little influence in the process.

The political leaders in the ministry scored relatively high from an instrumental perspective. They wanted to increase control and coordination in the health sector through a new organization structure. They initiated a change, allocated resources to achieve it and had professional competence; for the task at hand. And they defined the reorganization in general or collective terms. In sum: important conditions were met so that the political leaders could use their formal position to control the process.

The conflict-coalition perspective also throws light on the process. The political leaders had to build a coalition for two reasons. First, they had to decide upon which organization model to choose. The solution was simple: they adopted a model proposed earlier, under different conditions, by the administrative leaders in the ministry. Second, because of a high potential for conflict they needed support against their opponents in the directorate, both in deciding upon and implementing a new organizational structure. The administrators were in a position to give the necessary support.

What are the lessons to be learned from the analysis of the reorganization process? First, the study shows that politically salient reorganization in government can receive a lot of attention from political leaders who usually have capacity problems. This may indicate a growing concern with organizational design among political leaders (Olsen 1986).

Second, reorganization processes are often complex; the institutional milieu is heterogeneous as reflected in different views on reorganization, complex patterns of participants, problems, solutions, and choice opportunities (March & Olsen 1976). Under such conditions, political leaders have to

search

for alliance partners to support and implement their decisions.

Our

study illustrates that a coalition between political leaders and administrators can weaken the influence of professions and civil servant organizations in decision-making processes. This may be an element in a 're-hierarchical' trend in public policy (Olsen 1986).

Third, political initiative and coalition building are not enough to succeed. The definition of a reorganization is very important because it legitimates different ways of organizing the decision-making processes and thereby different decisions. This illustrates that there are competing decision procedures and norms in public organizations (Egeberg 1985; Egeberg & Stigen 1985). By emphasizing the 'majority rule' principle, the political leaders 'lifted' the reorganization process upwards in the hierarchy (Jacobsson 1984).

Our theoretical guiding line in analyzing reorganization effects has been the instrumental one. According to this perspective, a restructuring of public organizations on the basis of collectively approved goals will substantially changes or reshape the decision-making behaviour of the people concerned.

The two main indicators for changes in decision-making behavior were
changes in patterns of contact and working constraints.

The first hypothesis stated that a restructuring really matters and this was clearly confirmed by the analysis. A major part of the respondents reported some substantial effects of the reorganization on their patterns of contact and administrative milieu.

Our second hypothesis stressed that people who stayed in the health directorate through the reorganization, and thereby were affected by more comprehensive structural change, would experience more profound effects than people from the health department. This was also confirmed by the data. The prediction that respondents from the directorate would experience a weakening of their patterns of contact and a more problematic and turbulent administrative milieu was also fulfilled.

Last, we assumed that a reorganization differentiating between people in relation to structural changes would also give differentiated effects on decision-making behavior inside the two organizational units concerned. This too was confirmed by some respondents in both the directorate and the department who hold the view that they were marginally affected by the reorganization.⁸

On the basis of the individual decision-making effects mentioned above, what are the aggregated effects of the reorganization? One point of view is that the dramatic and differently experienced effects on contacts and administrative milieu show that the reorganization has been successful. The intention behind the reorganization was to strengthen the central political coordination in the health sector. This was attained by a successful health department

supporting the political leaders, a weakening and isolation of the directorate
in the decision-making processes, and an internal reshuffling in the directorate
creating a more clearly professional service unit.

From another point of view the reorganization was too comprehensive and a disaster for the health sector. The health directorate, a guarantee for a strong and professional health sector, was destroyed and isolated through the reorganization. The professional competence of the medical staff was decoupled from decision-making processes, the professional administrative milieu divided, and authority relations obscured. The changes in the roles of employees in the directorate were so profound that a lot of people did not manage to adapt to the new working conditions.

The different views on the aggregated effects symbolize a major dilemma facing actors in reorganization processes. In this case the political and administrative leaders in the ministry had to choose between political and cognitive considerations, between action and organizational thinking. They chose action, thereby leaving no room for influence from opponents of the; new organizational model. This strategy, paying less attention to the organizational principles behind the reorganization created legitimacy problems, especially in the directorate. Furthermore, there was some probability that ambiguous traits in organizational thinking could produce other effects than intended.

If, on the other hand, the leaders had chosen to use more time on a broad participation of different groups and on elaboration of different organization models, before the final reorganization decision was made, legitimacy could have been improved, but probably another model would then have been chosen and the political control weakened.

---

NOTES

1. By 'political elites' we refer to members of the cabinet and other political leaders in the ministries. Studies show that parliaments are not heavily involved in administrative reform.

2. Directorates, as central administrative units connected with ministries, have historically had three different forms: independent (the most common one), semi-autonomous, and dependent (placed in the ministry as a department).

3. The people in the health department already were in the ministry before the reorganisation, but in a semi-autonomous position. This change must be less dramatic than 'leaving' the ministry as the people in the new directorate did.

4. The focus is on the actors' reactions to what they have experienced after a reorganization has taken place (Egeberg 1984). It is the individual actor who feels the effects of reorganization and changes his/her decision-making behaviour. The 'sum' of individual reactions constitutes the aggregated effects for the whole institution. When people in the administration act they are assumed to be influenced by rather simple models of thought constructed on the basis of a selection of facts (March & Simon 1958; Simon 1957; March & Olsen 1976). Actors' opinions reflect their models of thought, their definitions of problems and solutions and thereby presumably systematic traits in their actual decision-making behaviour (Laegreid & Olsen 1978; March & Olsen 1976).

---

5. The number of respondents from the directorate and the department are 32 and 9. The majority of the se people are administrators, not executive officers. 81% of the administrators in the directorate have been interviewed, 70% in the department. Since the number of respondents from the department are rather small we place the percentages in brackets and are careful in drawing conclusions from these data in the analysis. Therefore we comment more on the data from the directorate.

6. It is difficult to find a specific measure on formal structural change. We use one indicator basedon the number of former colleagues gathered in the same unit in the new structure. The actors who through the reorganization have lost most of their former colleagues score high on formal structural change. This is, for example, 50% of the respondents from the directorate.

7. 73% of the respondents from the directorate scoring high on formal structural change report major effects on their contacts and administrative milieu.

8. 59% of the respondents from the directorate scoring low on formal structural change report minor effects on their contacts and administrative milieu.

REFERENCES

Christensen, T. 1983. Hovedtrekk i utviklingen av direktoratene som organisasjonsform i norsk
sentralforvaltning - endringer i problemoppfatninger og løsningsforslag. Institute of Social
Sciences, University of Tromsø.

Christensen, T. 1985. Styrt endring og planlagte konsekvenser? -en studie av omorganiseringen
av den sentrale helseforvaltningen i 1983. Institute of Social Sciences, University of Tromsø.

Davis, T.R.V. 1984. 'The Influence of the Physical Environment in Offices', Academy of Management
Review, 9, 271-83.

Egeberg, M. 1984. Organisasjonsutforming i offentlig virksomhet. Oslo: Aschehoug/Tanum-Norli.

Egeberg, M. 1985. 'Designing Public Organizations'. To be published in Eliassen & Kooiman,
eds.; Public Management. Sage, 1987.

Egeberg, M. & Stigen, I. 1985. 'The Management of Competing Norms and Decision Principles:
The Organizational Context of Norwegian Directorates'. Paper prepared for presentation at the
IPSA XHlth World Congress, Paris. 15-20 July 1985.

Eriksen, S. 1986. Samordningsproblemer og samordningstiltak. Britiske erfaringer med sammenslåing
av departementer. Oslo: FAD/RFSP.

Garnett, J.L. & Levine, C.H. 1980. 'State Executive Branch Reorganization. Patterns and Perspectives',
Administration and Society 12, 227-276.

Goodsell, C. 1977. 'Bureaucratic Manipulation of Physical Symbols: An Empirical Study', American
Journal of Political Science, 21, 79-91.

Grafton, C. 1979. 'The Reorganization of Federal Agencies'. Administration and Society 10,
437-464.

Grønnegård Christensen, J. 1982. 'Den administrative ledelsesfunktion i centraladministrationen',
Nordisk Administrativt Tidsskrift, nr. 4.

Gulick, L. 1969. 'Notes on the Theory of Organization. With Special Reference to Government',
in Gulick, L. & Urwin, L. eds. Papers on the Science of Administration. N.Y: A.M. Kelley.
(First edition 1937).

Hood, C. 1979. "The Machinery of Government Problem'. Studies in Public Policy. No. 28.
Glasgow: University of Stratchclyde.

Jacobsson, B. 1984. Hur styrs forvaltningen? - myt och verklighet kring departementens styrning
av åmbetsverken. Stockholm: EFI/HHS.

Lægreid, P. & Olsen, J.P. 1978. Byråkrati og beslutninger. Oslo: Norwegian University Press.

March, J.G. & Olsen, J.P. 1976. Ambiguity and Choice in Organizations. Oslo: Norwegian Univertisy
Press.

March, J.G. & Olsen, J.P. 1983. 'Organizing Political Life: What Administrative Reorganization
Tells Us About Government', American Political Science Review 77, 281-97.

March, J.G. & Simon, H.A. 1958. Organizations. London: Wiley.

Mayntz, R. & Scharpf, F. 1975. Policy-Making in the German Federal Bureaucracy. Elsevier.

Olsen, J.P. 1981. 'Integrated Organizational Participation in Government', in Nystrom, P.G. &
Starbuck, W.H., eds, Handbook of Organizational Design. Vol. 2, 492-516. New York: Oxford
University Press.

Olsen, J.P. 1983. Organized Democracy. Oslo: Norwegian University Press.

Olsen, J.P. 1986. 'Administrative Reform and Theories of Organization'. To appear in Campbell,
C. & Peters, 8.G., eds., Organizing Government. Pittsburgh: University of Pittsburgh Press.

Pfeffer, J. 1982. Organizations and Organization Theory. Pitman.

Pollitt, C. 1984. Manipulating the Machine. Changing the Pattern of Ministerial Departments.
1960-83. London: George Allen & Urwin.

Roness, P.G. 1982. 'Organisering av endringsprosesser i den norske sentraladministrasjon', Unpublished
manuscript, Institute of Public Administration and Organisation Theory, University of
Bergen.

Salamon, L.M. 1981. 'The Question of Goals', in P. Szanton, ed. Federal Reorganization: What
Have we Learned? N.J. Chatham.

Scott, W.R. 1981. Organizations. Rational, Natural and Open Systems. Englewood Cliffs, N.J.:
Prentice Hall.

Seidman, H. 1980. Politics, Position and Power. The Dynamics of Federal Reorganization. Oxford
University Press. 3rd. edition.

Simon, H.A. 1957. Administrative Behaviour. Free Press.

Weber, M. 1970. 'Bureaucracy', in Gerth, H.H. & Mills, C.W, eds., From Max Weber. London
Routledge and Kegan Paul.