Political leaders and public administrators: Interaction patterns and pictures in Finnish local government decision-making processes

Abstract

The rapid change of local government operating environment shapes the interaction between political leaders and public administrators, who work in the constant riptide of service responsiveness and economic pressure. We investigate the relationship between political leaders and public administrators in the local administration of social and health services. The patterns and pictures are examined empirically, with data gathered from strategic-level political leaders and public administrators in six Finnish local government organisations. The analysis applies multivariate methods. The results suggest that there are different groups among the political leaders. The differences are not based on political opinions, but rather on the attitudes towards the decision-making process, also the views on local government decision-making processes differ between the groups. The rapid contextual change experienced in the public organisations requires a fast and well-informed response from political leaders and public administrators; even political differences can be pushed aside in this turbulent operating environment.

Keywords

Decision-making interaction Finnish local government political leaders public administrators

Introduction

The politics-administration dichotomy is one of the cornerstones of modern Western democracies (Hansen and Ejersbo, 2002). Similarly, the relationship between political leaders and public administrators interests researchers of both political science and public administration. Woodrow Wilson (1887) illustrated the dichotomy already at the end of the 19th-century, and the relationship was developed into an ideal type model by Max Weber (1922). Max Weber’s classic model of bureaucracy is a simplified theoretical illustration of the roles of political leaders and public administrators, in which political leaders represent the people’s opinions, whereas the public administrators administer impartially (Mouritzen and Svara, 2002). In Western democracies, this ideal type model is an important context factor in public sector strategic management and decision-making (Ring and Perry, 1985), and the relationship between the actors in the two spheres is highly interesting as the interaction between them continuously shapes the relationship between the spheres.

Local government and its role, powers, functions and constitutional status in the European countries have been shaped by the diverse historical, cultural, social and governmental systems and traditions, the politics-administration dichotomy being one of the important characteristics. In Finland and other Nordic countries, local government is responsible for a large share of public services, ranging from social and health care to education and infrastructure. At the same time, local government is one of the most important arenas of participation in democratic life (Local Government Act 27§ 519/2007). In the middle of the 2010s, Finland is undergoing a major public sector structural reform, which has to do with the interaction and operations models in the local government. However, the reform does not remove the old layers of administration, which persist as ways of acting in the organisation.

Political participation has become more individualised, ad hoc, issue specific and less linked to traditional societal cleavages than before (Horvath and Paolini, 2013) and, at the same time, the demands for direct citizen participation are expressed more often than before (Dalton, 2008; Sloam, 2013). The ways people participate and communicate are changing, which exerts pressure on the traditional political decision-making process based on classic representative democratic theory. These developments may indicate that the new generation of political leaders and public administrators has new ways of acting in the local government decision-making process (Niiranen and Joensuu, 2014).

Local government organisations usually have a long historical background with institutionalised (Walgenbach and Meyer, 2008: 55–63), fossilised (Guba and Lincoln, 1994) or slow structures which have developed over a long period of time. With today’s fast paced environment and global effects, the organisations have historical layers with extremely different characteristics. These layers are mirrored in the interaction between political leaders and public administrators, especially in the decision-making processes (Niiranen and Joensuu, 2014). Understanding the multi-layered and multifaceted system entails studying the continuous organising and interaction in the organisation (Follett, 1973: 166; Harisalo, 2008: 84–85; Keski-Petäjä, 2009: 67). Notably, interaction is not only a product of today’s operating environment, but also characterised by the local historical and political culture. For example, there seem to be differences between the political leaders and public administrators in how they perceive the information used in decision-making. In a simplified manner, the public administrators think that the political leaders mostly use the information they provide on the agenda in the decision-making process. However, the political leaders consider the information to be a process, which includes both the agenda and collects other information from written sources as well as in interaction with the municipal residents (Niiranen et al., 2013).

The purpose of this paper is to investigate the interaction patterns and pictures of the political leaders and public administrators in Finnish local government. We are interested in the interaction between two key groups of strategic managers: political leaders and public administrators, and try to discern interaction patterns within the group of political leaders and within the group of public administrators. The study design and analysis are based on Margaret Archer’s (1995: 132, 165–194) theorising on the relationship between structure and agency. The organisations studied are reforming and, according to Archer, the structural elaboration and the reproduction or transformation of the initial structure are dependent on the existing structures. At any particular moment, the existing structures (like organisations, decision-making procedures and the larger societal context) constrain and enable agents, whose interactions produce both intended and unintended consequences. These different ways of interacting may have an effect on how the political leaders and public administrators work together and, more generally, how they picture the local government decision-making process.

Research questions

Our study investigates the relationship between political leaders and strategic-level public administrators in the local administration of social and health services in Finland. Our research questions are:

Do the political leaders and public administrators have different kinds of interaction patterns and, if so, what kinds of patterns are they?

What effect do these possible patterns have on the ways in which the political leaders and public administrators work?

What kind of views do the strategic-level political leaders and public administrators have on the decision-making process, do the views differ from each other and, if so, why?

The first research question is addressed by condensing, by means of exploratory factor analysis, the research data on how political leaders assess the decision-making process in their municipalities. On the basis of this assessment, the actual observations are then clustered according to the respondents’ experience of the political decision-making process. The experience of the local government decision-making process may indicate the interaction pattern of the group of political leaders. Public administrators were divided into two groups based on their tasks as a presenting or non-presenting public administrator. To answer the second research question, the data were divided according to the clusters and then contrasted with data sets on the political leaders’ work and cooperation between political leaders and public administrators. The third question addresses the views of the decision-making process. These were investigated with a part (five statements) of one data set measuring the view in 19 dimensions. In the analysis, the groups of public administrators and political leaders were treated as separate entities.

Focus and theoretical starting points of the study

Finnish local government system in brief

In Finland, local government (municipalities) is separate from central government, and the municipal bodies are partly independent of the state. The core of the local government, the municipalities, comprises self-governing entities, which, under Finnish law (the Constitution of Finland 597/2007 and the Local Government Act 519/2007), have the right to levy taxes and to decide how to organise their functions. Local self-government is the foundation of political decision-making, municipalities must adhere to democratic principles and the local elections are based on universal suffrage. The Local Government Act stipulates how municipalities may organise their administration and recognises their diversity, as external and internal conditions vary widely in different parts of the country. Municipalities are responsible for providing their residents with statutory basic services, for example, education and culture, social welfare and health, technical infrastructure, and the environment (State Treasury, 2015; Local Government Act 519/2007 §1). The major local government reforms connected with the government programmes and structural reform of the municipalities started in 2007 (Kettunen, 2015). These reforms constituted one layer of the local government operating environment during the data collection of this study, but the reforms seem to be intensifying at the time of writing.

All municipalities must have a municipal council, board, audit committee, and an election committee responsible for organising and holding elections every four years. A municipal manager, who is not a member of the local council, must also be appointed (Local Government Act 519/2007). The municipal council is the supreme decision-making body, and its members are elected for a four-year term in local elections. The council elects members to the municipal board, which is responsible for preparing matters for the council and implementing its decisions. The municipal committees are in charge of providing local public services and the committee members are elected by the municipal council (State Treasury, 2015). Committee members do not need to be members of the municipal council. The items on the municipal committees’ agendas are prepared by the public administrators, and presented at the meetings by a presenting public administrator, usually the managing public administrator in the sector (Kurikka, 2014).

The relationship between politics and administration

In Finland, as in many other countries, the local government system is based on the previously explained Weberian notions of a bureaucratic ideal model (Weber, 1922). The attempt to maintain distance between the roles of political leaders and public administrators is a phenomenon as old as separating politics and administration. However, research has shown that politics plays a role in administration (Long, 1949; Simon et al., 1950), and that the relationship between the two groups is much more complicated than the ideal type model suggests. For example, James H. Svara (1998, 1999) is opposed to the simple interpretation of clear roles, and has developed a complementarity model of politics and administration. Many recent researchers consider complementarity or other similar approaches to be a more appropriate conceptualisation than dichotomy in understanding political–administrative relations (Demir and Reddick, 2012). Peters and Pierre (2003) suggest that politics and administration should be thought of as different elements of the same process of formulating and implementing policy.

In the 1980s, researchers noticed that decision-making environments had become increasingly complex, which then necessitated a lot of work, capacity, and authority from both political leaders and public administrators (see e.g. Olsen, 1983). In the local government context, the change has both external reasons, such as structural changes, the global economic crisis, increasing service needs and ageing, as well as internal reasons, such as local economic challenges and continuous reforms (Berg and Rao, 2005; Denters and Rose, 2005). Many local government organisations address the constant change and pressure by reforming and reshaping their organisational structures and processes, and the public sector in most Western countries has experienced continuous reforms since the late 1970s (Jacobsen, 2005). In Finland, these reforms have taken different forms, but in the 2000s governments have been actively promoting and planning reforms in municipalities and the social and health care sector, which is largely a responsibility of municipalities (Kettunen, 2015). Implementing the reforms depends on the characteristics of the governmental system of the country in question (Pollitt et al., 2007), and the reforms in public organisations have initiated transformations in the relations between political leaders and public administrators, but to varying extents, depending on how the reforms have been conducted in the countries in question (Giauque et al., 2009; Pollitt and Bouckaert, 2004).

The contemporary problems in local government across Europe are very similar, but the means, methods and possibilities for solving these vary because of the different institutional arrangements for addressing policy, political, economic and social issues (Copus, 2012). In the Finnish context, the same phenomenon is visible on the local level, as the municipalities solve similar problems in different ways depending on the operating environment of the municipality (Niiranen and Joensuu, 2014; Wynn and Williams, 2012: 797). The fast pace of change, continuous reforms, global, national and demographic issues and technological development are intertwined in the local government operating environment of the 2010s, and these aspects shape the interaction between political leaders and public administrators at the strategic level, and they generate new kinds of interaction between the political leaders and public administrators.

The turbulent operating environment seems to require more participation, transparency, dialogue and faster decisions than the current local government decision-making system in Finland can achieve (Strandman, 2009: 215). The political leaders in the midst of continuous change are using their extensive networks, strong democratic beliefs and global connections more and more (Szücs and Strömberg, 2009). The turbulence makes political decision-making difficult because wisdom and dexterity are needed in situations where there are contrasting issues or issues in which the consequences are difficult or impossible to predict (Rapeli, 2010). Sometimes the political leaders’ authority in local government has been considered to be minimal, whereas the public administrators have been considered to run the municipality. However, the turbulence seems to call for stronger leadership at the same time as leadership has been politically restricted in scope (Verheul and Schaap, 2010: 451).

The relationship between political leaders and public administrators is constructed in social interaction and is continuously reshaped (Bergström et al., 2008: 211). The operating environment, other actors, traditions or culture have an effect on the actual interaction between political leaders and public administrators, but individual ways of interacting are also important (Figure 1). The political leaders also have an important role in contributing to the formation of a community’s identity. This is important, as it enables the political leaders to keep their communities together in an increasingly complex environment (Verheul and Schaap, 2010: 451).

Figure 1.

Interplay of structure and agency in local government.

Theoretical perspectives on public decision-making processes

Decision-making in politics is much more than the date and time when the decision is formally made. It is a social process involving many people who have to interact over prolonged periods of time to reach semi-consensual decisions. The process involves various sub-processes of technical problem-solving, social adjustment, political bargaining, negotiations and organisation-wide communications (Shrivastava and Nachman, 1989). In the Finnish local government organisations, these processes involve both political leaders and public administrators on different levels of the organisations. Interaction with municipal residents is also important. Managing a municipality, or another kind of local government organisation, is not a technical-rational function based on regularities, but rather a social and political phenomenon (McAuley et al., 2007: 113). Thus, political decision-making depends more on group processes and exchanges transcending departmental and organisational boundaries than on behaviour or interactions among a few key individuals (Mason and Mitroff, 1981; Mintzberg et al. 1976; Quinn 1980; Shrivastava and Nachman, 1989).

Public decision-making processes and the interaction they entail have been theoretically described from a number of perspectives (Christensen and Lægreid, 2002). According to the structural perspective, the formal structure of the public system is the most influential factor in public decision-making processes (Gulick, 1937; Simon, 1957). The interaction is based on hierarchy, specialisation or collegial principles. The cultural-institutional perspective regards the informal norms as the basis of the interaction. The development of cultural traditions in politico-administrative systems is path-dependent and gradual. Issues of trust and conflict become important because the relationship between political leaders and public administrators is based more on informal interaction than on formal control and decisions (Selznick, 1957). The exposure perspective underlines the increased complexity and unpredictability, and the related capacity problems of the political leaders and public administrators (March, 1981). Time, energy and attention are scarce resources, and the interest is in how the political leaders and public administrators perform their duties in these conditions (Christensen and Lægreid, 2002; Lægreid and Roness, 1999).

Most of the research on the interaction between political leaders and public administrators has been conducted on the national level (central government). The interaction between political leaders and public administrators works in a similar manner both in central and local government. However, there are some differences between the levels, as in Finland on the local level there is no formal opposition. Furthermore, on the central level the political leaders work full time in politics whereas on the local level most of the political leaders have full-time employment elsewhere. The linkages between central and local levels are close; there are, for example, local political leaders who have positions on the central level as well.

Research data

Research data were gathered from six Finnish local government organisations in 2011 and 2012 using two questionnaires (N = 202). The six organisations¹ from which the data were gathered represent a cross-section of Finnish municipalities. They have many features in common, but differ in size, location and the way social and health-care services are organised. Four of the municipalities are rather large by Finnish standards; one is middle-sized and one is a co-operation district combined of three small municipalities (see e.g. Kettunen, 2015: 56). Most inhabitants in the four largest local government organisations live in densely populated areas, the smaller ones are rural. All organisations have carried out some type of major reform during the five years before data collection; they have implemented both municipal mergers and other kinds of organisational reforms. This reflects the situation in the Finnish municipal sector well in general and provides a platform for studying the interaction between political leaders and public administrators in reforming local government organisations.

We used qualitative interview data (N = 27) as the basis for the quantitative research. The data were gathered from chairpersons of municipal councils, executive boards, the committees responsible for social and health-care services and members of boards of directors responsible for producing social and health care services (N = 18). The data were also gathered from strategic-level leaders or middle management in the central administration of social and health-care services (N = 9). The interviews were transcribed and examined using theory-dependent content analysis (Molina-Azorin, 2012). After reading the transcribed texts, the findings were classified with an analysis framework into themes, which provided the basis for the themes addressed in the questionnaires. The themes in the two questionnaires were political leadership and value base, knowledge base and interaction. The questions were mostly formulated differently for political leaders and for public administrators to capture the views of the groups from their perspectives as well as possible. The questionnaires were pre-tested for content validity by groups of political leaders and public administrators not belonging to the actual group of respondents. After pre-testing, minor changes to enhance the clarity of the questionnaires were made. The respondents to the questionnaires were political leaders and strategic administrators of the case organisations.

The questionnaire for political leaders was sent to the municipal councils, executive boards and committees responsible for social and health care in the six case organisations (N = 459). The response rate after four reminders was 36.2% (n = 166) and all the returned questionnaires were usable. Unit nonresponse was analysed regarding organisation, political affiliation and gender of the respondent. The response rate between organisations varied from 32% to 38%. The respondent’s political affiliations, compared to the political affiliations in the target group, were nearly equal, and the respondents represent well the political map of the organisations. The gender balance of the respondents represents well the gender balance of the political leaders the questionnaire was sent to. Item nonresponse was analysed question by question, and the largest nonresponse was encountered in the open questions. In the questions analysed for this article, the item nonresponse varied between 0% and 1.8%.

The questionnaire for the public administrators was sent to representatives of strategic and middle management in the social and health care sector in the case municipalities (N = 65). The response rate after three reminders was 56.9% (n = 37) and all returned questionnaires were usable. Unit nonresponse was analysed regarding the organisational affiliation and gender of the respondent. The largest unit nonresponse was in an organisation which encountered political turbulence at the time of the data collection. The smallest unit nonresponse was in the organisation with only minor reforms prior to data collection. Men were slightly overrepresented among the respondents. Item nonresponse was most common regarding the open questions, but it was almost non-existent in the other questions. In the questions analysed for this article there was no item nonresponse. Thus, we consider that the respondents represent the target group sufficiently, but a higher response rate would have been desirable (Baruch and Holtom, 2008).

Analysis method and results

Patterns of interaction within the group of political leaders

The first research question is addressed using exploratory factor analysis to condense the research data on how political leaders assess the decision-making process in their municipalities. Based on the assessment, the actual observations were then clustered to characterise the interaction patterns of the political leaders. To answer the second research question, the data were divided according to the clusters and then contrasted with data sets on the political leaders’ work and cooperation between political leaders and public administrators.

The political leaders assessed a series of 19 statements on how much the issues mentioned influenced the decision-making process in their respective municipalities. The analysis was conducted using SPSS statistical package. An exploratory factor analysis on the political leaders’ opinions on the municipal decision-making process was conducted to find the underlying structure in the dataset (Hair et al., 2014: 97). Exploratory factor analysis is based on the fundamental assumption that some underlying factors, which are smaller in number than the observed variables, are responsible for the covariation among them. The purpose of the analysis is to find the smallest number of interpretable factors which can adequately explain the correlations among a set of variables. Variables, which load on the same factor, measure the same underlying construct (Kerlinger, 1986: 19). This fits well with the critical realist starting point of the study, as the structure of the different aspects of the decision-making process becomes visible. We had preliminary ideas based on previous research about the possible factors, so we were interested in analysing only the variance that is accounted for instead of analysing all possible variance between the variables (Tabachnick and Fidell, 2007, 635).

Principal axis factoring was chosen as the analysis method instead of the maximum likelihood method as the questionnaire answers measuring opinions on a Likert-scale are not normally distributed and, in that sense, the assumption of multivariate normality is violated (Fabrigar et al., 1999). When considering the choice of rotation method, there is reason to assume that opinions measured on a Likert-scale correlate with each other, so an oblique rotation can be considered to be theoretically more valid than an orthogonal rotation. The oblique solution also provides information on the correlations between the factors (Hair et al., 2014: 112–113).

In the first phase of the analysis, the communalities of the variables were analysed and the variable ‘Availability of workforce’ was dropped from the analysis due to an extremely low communality (0.12), which can be considered to be evidence that it has little in common with the other variables analysed, and thus is not a reliable measure of the proposed factor solution. After that the analysis was re-run with the remaining 18 statements, which loaded onto four factors with eigenvalues greater than one. However, a solution with five factors was accepted after reviewing the eigenvalues, the number of non-trivial factors,² and experimenting with the proposed solutions to find the simplest³ factor solution. The extraction sums of squared loadings accounted for 59.3% of the explained variance, which can be considered satisfactory as the factors correlate with each other (Hair et al., 2014: 107–108). The set of issues influencing decision-making in the municipalities was reduced to five underlying factors as shown in Table 1.

Table 1.

A five factor solution of the question ‘How much do the following issues influence the decision-making process in your municipality? Assess on a scale from 1 to 6 representing 1 = Not at all, 2 = A little, 3 = Somewhat, 4 = A lot, 5 = Very much and 6 = Cannot say’ (KMO .857, Bartlett’s .000).

Pattern matrix^a
	Factor
	1	2	3	4	5	h²
Lack of trust between public administrators	.616					.569
Lack of trust between political leaders	.559		−.389			.665
Lack of trust between public administrators and political leaders	.479		−.303			.681
Municipal residents’ opinions on municipal decision-making		.932				.848
Municipal residents’ feedback on the service		.853				.750
Initiatives presented by municipal residents		.677				.453
Prejudice against other political leaders (negative)			−.834			.775
Prejudice against subject matter (negative)			−.764			.625
Prejudice against public administrators (negative)			−.654			.602
It is hard to obtain information or there is no information available				.853		.817
Communication between the different actors is poor				.563		.543
Preparation of the subject matter is insufficient				.491		.615
The short-term nature and inconsistency of politics					.624	.629
Decisions and strategy do not meet in our municipality					.624	.621
Concentrating on details instead of the big picture			−.338		.522	.492
Economic situation in the municipality					.500	.226
Political leaders have not enough time to acquaint themselves with matters to be decided					.489	.411
Some of the political leaders in my municipality hold strictly to their own political outlooks					.353	.368
Eigenvalue	6.62	2.47	1.38	1.19	.95
Proportion of variance	34.6	12.0	5.7	4.2	2.8

Extraction Method: Principal Axis Factoring; Rotation Method: Oblimin with Kaiser Normalisation.

Communalities lower than 0.3 are suppressed from the table for the clarity.

Rotation converged in 16 iterations. Boldface highlights the values included in each factor.

The factors were examined and given a descriptive title that represented the characteristics of the constructs. The five factors can be considered to be dimensions influencing the local government decision-making process. The first factor, lack of trust, contained three variables that addressed the lack of trust between different actors in the local government decision-making process. The second factor, municipal residents, included three forms of municipal residents’ responses to local government decision-making. The third factor, prejudice, consists of three variables measuring prejudices against people and subject matter. The fourth factor, information process, describes the whole process of obtaining information from preparing subject matter to communication. The fifth factor is called strategic soundness, and encompasses six variables. In the further analysis, the summated scales based on the factor analysis were used. The summated scales were tested using Cronbach’s alpha, which in all cases had values above 0.79, meaning that the summated scales can be considered reliable.

By clustering the data on political leaders we wanted to explore into what kind of groups the respondents could be divided based on the five dimensions of the decision-making process. The aim was to create groups which were as homogeneous as possible. The initiative clusters were created using Ward’s method and the analysis was amplified using k-means clustering (Hair et al., 2014: 446–448). A three-cluster solution was chosen and the groups were named according to the dimensions they represented by studying the means and mean centred values of the grouping dimensions (Table 2). The first group is trustful, and the group members are more positive in their attitudes towards all the grouping variables except the municipal residents variable, in which there is no difference between the three groups. The second group is called middle-of-the-road, and includes about half of the respondents. The group members are middle range on all clustering dimensions. The third group is critical, in which the group members have more negative attitudes towards the grouping variables than the middle-of-the-road group. The differences between the groups are statistically significant on all the grouping variables other than municipal residents (p < 0.001).

Table 2.

Clusters, cluster names and cluster centres (N = 166, missing 3)

Cluster	n	Percentage	Cluster name	Lack of trust	Municipal residents	Prejudice	Information process	Strategic soundness
1	46	28.2	Trustful	1.92	2.61	2.01	2.15	2.76
2	85	52.1	Middle-of- the-road	2.82	2.95	3.10	2.86	3.63
3	32	19.6	Critical	4.07	2.58	3.90	4.07	4.34

The group members were profiled with the background variables age, education, party membership, occupational group, domicile, working in the same municipality where elected as a political leader and length of service on the municipal council. There was no difference in age, education or gender profiles between the groups. The critical attitude group differed from the two other groups as the members were more often working in the same municipality in which they had been elected as political leaders (38% vs. 25% in the other two groups). Likewise, the members of the critical group were more often retired or otherwise not in working life (41% vs. 33% middle-of-the-road group and 22% trustful group). There were also slight differences in the domiciles between the three groups. The members of the middle-of-the-road group lived equally often in the centre, suburbs and sparsely populated areas. The members of the trustful group lived slightly more often in the centre (41%) than the members of the other two groups, whereas the members of the critical group lived slightly more often in the sparsely populated areas (41%) than the members of the other two groups. For those who live far away from the centre, the distance to service is usually greater, which may affect their view on both service production and service quality. In addition, because of the distance meeting public administrators or other political leaders unofficially outside the committee, board or council meetings requires more time and effort and may be more difficult. The differences in political affiliations between the groups were very small, but some differences were found between the groups according to length of service on the municipal council. In the trustful group, 69% had been members of the municipal council for more than 11 years, whereas, in the other two groups, approximately 50% of the group members had been council members for more than 11 years.

The effect of the interaction patterns on political leaders’ work was studied by comparing the opinions on six statements measured on a five-point Likert scale, which was condensed into a three-point scale for purposes of analysis and clarity (Table 3). In two of the statements, the differences between the three groups were statistically significant.

Table 3.

Political leaders’ responses to statements on cooperation between political leaders and public administrators and on the work of a political leader

	Trustful (n = 46)	Middle-of-the-road (n = 85)	Critical (n = 36)	All (n = 163)
Personal relations have an effect on decision-making in our municipality (N = 161)^a
Agree	20%	53%	77%	48%
Neutral	18%	22%	16%	20%
Do not agree	62%	25%	7%	32%
	100%	100%	100%	100%
In my position, I often have to balance between the economy and my values (N = 162)
Agree	83%	87%	97%	88%
Neutral	4%	9%	0%	6%
Do not agree	13%	4%	3%	6%
	100%	100%	100%	100%
In the decision-making discussion of my municipal organ it is usual to consider various possible solutions (N = 162)
Agree	48%	43%	56%	41%
Neutral	24%	14%	16%	17%
Do not agree	28%	43%	28%	41%
	100%	100%	100%	100%
The real decision-making power in the municipalities rests with the public administrators, not the political leaders (N = 159)^b
Agree	51%	65%	80%	64%
Neutral	4%	10%	3%	7%
Do not agree	44%	25%	16%	29%
	100%	100%	100%	100%
Presenting public administrators use the political leaders’ expertise (N = 160)
Agree	54%	47%	34%	47%
Neutral	14%	17%	6%	14%
Do not agree	32%	36%	60%	39%
	100%	100%	100%	100%
Preparation seminars are most often a waste of time for the political leaders (N = 158)^a
Agree	4%	21%	39%	24%
Neutral	7%	17%	3%	11%
Do not agree	89%	62%	58%	65%
	100%	100%	100%	100%

Statistically significant difference between groups, p < 0.001, χ²-test.

Statistically significant difference between groups, p < 0.05, χ²-test.

The difference in opinions between the trustful group and the critical group on personal relations having an effect on decision-making was large and statistically significant. The trustful group was polarised in its view on decision-making power in municipalities, the percentages of those who agreed and those who disagreed were almost equal. The majority (80%) of the critical group reported that the public administrators wielded the real power in the municipalities, which may indicate problems in the relationship between the two groups. To confirm this kind of relationship between the two statements would require further analyses. However, it is very interesting that almost two-thirds of all the respondents considered the public administrators to have the real decision-making power. It is possible that political leaders consider that a lot of the power to be vested in the preparation phase of decision-making process because public administrators are trained experts on the subject matters, whereas political leaders may have another field of expertise.

The critical group reported slightly more often than the other two groups that different possible solutions were considered in the decision-making discussion. It is possible that the group was more critical in the discussion and brought new views to the table. As political leaders’ task is to allocate values (Askim, 2007), it is understandable that balancing between economy and one’s own values is a key element for every group. It is possibly also an indication of a feeling of responsibility in their role as political leaders and their accountability towards municipal residents. Approximately half of the trustful and the middle-of-the-road groups agreed on the public administrators using the political leaders’ expertise whereas 60% of the critical group did not agree with the statement. Could it be that these political leaders thought that they would have more to give if only it were used? The preparation seminars were much appreciated in all groups. However, the opinions in the critical group were polarised, as almost 40% of the group reported that the seminars were waste of time for the political leaders. In general it seems that the preparation seminars provided a discussion forum for the political leaders.

Patterns of interaction within the group of public administrators

The data set on public administrators is small (N = 37) even if the response rate was sufficient (56.9%). The group of public administrators was divided into two categories, the presenting public administrators and the public administrators whose work does not include the presentation of items on the agenda. The basis for this grouping is totally different from that of the political leaders as it is based on the tasks of the public administrators, not their views on the interaction in the decision-making process. In the Finnish municipal system, items on the agenda must be presented to the committee or board by a designated public administrator (usually the strategic manager of the field in question), hence the concept ‘presenting administrator’. This division is, therefore, interesting as the presenting public administrators co-operate more closely with the political leaders than do the public administrators. The other public managers often prepare items on the agenda, but the presenting administrators work in closest cooperation with the political leaders and interact more with them. There were 23 respondents in the group of presenting public administrators and 13 respondents in the group of non-presenting public administrators (N = 37, one missing).

The effect of the interaction patterns of presenting and not presenting public administrators’ work was studied by comparing the opinions on six statements measured on a five-point Likert scale, which was condensed into a three-point scale for the purposes of analysis and clarity. None of the statements revealed statistically significant differences between the groups, but practical differences became apparent in the percentages (Table 4). A larger data set would possibly also have yielded some statistically significant differences between the groups.

Table 4.

Public administrators’ responses to statements on cooperation between political leaders and public administrators and on the work of political leader

	Presenting	Non Presenting	All
Personal relations have an effect on decision-making in our municipality (N = 35)
Agree	22%	50%	31%
Neutral	9%	25%	14%
Do not agree	69%	25%	54%
	100%	100%	100%
In my position as a presenting administrator I often have to balance between economy and ethical values (N = 32)
Agree	70%	45%	62%
Neutral	9%	33%	16%
Do not agree	22%	22%	22%
	100%	100%	100%
As a presenting administrator I present different possible solutions in the decision-making discussion (N = 30)
Agree	74%	57%	70%
Neutral	4%	29%	10%
Do not agree	22%	14%	20%
	100%	100%	100%
The real decision-making power in municipalities rests with the political leaders, not the public administrators (N = 36)
Agree	57%	62%	58%
Neutral	26%	0%	17%
Do not agree	17%	38%	25%
	100%	100%	100%
In my work, I use the political leaders’ expertise (N = 35)
Agree	65%	33%	54%
Neutral	22%	25%	23%
Do not agree	13%	42%	23%
	100%	100%	100%
Preparation seminars are most often a waste of time for the political leaders (N = 35)
Agree	4%	0%	3%
Neutral	4%	8%	6%
Do not agree	92%	92%	91%
	100%	100%	100%

The difference in the position between a presenting public administrator and a non-presenting public administrator was apparent in their reactions to on the statements on cooperation between political leaders and public administrators and on the work of a political leader. The presenting public administrators seemed to view the decision-making more as a matter of agenda issues than personal relations than did the non-presenting public administrators as 70% of them did not agree with the first statement on personal relations. The contact between non-presenting public administrators and political leaders is less frequent and more remote than the contact between presenting public administrators and political leaders. Presenting public administrators discuss the subject matter with political leaders both unofficially, in seminars and in council, board or committee meetings. Thus, they can more easily assess the effect of personal relations in decision-making than non-presenting administrators. Balancing between ethical values and economy is an essential part of the work of presenting public administrators (70% agreed). They often have both overall and budgetary responsibility on their area of expertise, and they encounter the most problematic issues to be solved in their organisations. However, non-presenting public administrators also consider ethical balancing to be a central feature in their duties (45% agreed).

The presenting administrators offered different possible solutions for discussion very often (75% agreed) and they used the political leaders’ expertise when needed (65% agreed). This indicates that discussion with political leaders and their knowledge is valued. The responses of non-presenting political leaders (42% did not agree) may also point towards them interacting less with political leaders in general. Approximately 60% of all the responding public administrators reported that the political leaders wielded the real power in the municipality. However, the opinions of the non-presenting public administrators were clearly polarised, as about 40% of them did not agree with this. This could be a result of their not being in close contact with the political leaders in the decision-making process. Interestingly, 26% of the presenting public administrators were neutral in this question, which could mean that they see the real decision-making power to be shared. Like the political leaders, the public administrators also reported that the preparation seminars were useful for the political leaders.

Pictures of the local government decision-making process

The pictures of the decision-making process were investigated with five variables from one data set measuring the picture on 19 bipolar descriptions. Both the political leaders and public administrators rated their pictures of the decision-making process in their municipalities on a seven-point semantic differential rating scale for the following five bipolar items: unanimous decision – vote, based on information – based on emotions, public administrator-led – political leader-led, trust – distrust, and bureaucratic – flexible. However, due to the large difference in the size of the data sets, it is not feasible to compare the results, and should be treated as different elements of the same process of formulating and implementing policy (Peters and Pierre, 2003).

A one-way analysis of variance (ANOVA) of the bipolar picture descriptions was calculated for both political leaders (three subgroups) and public administrators (two subgroups). Concerning the political leaders the analysis was carried out between the groups trustful, middle-of-the-road and critical, and public administrators were divided in presenting and non-presenting administrators. All the analyses on the data on political leaders were significant (Table 5). Afterwards, Bonferroni post hoc (p < .05) tests were run on the data on political leaders to find the significant differences between the groups (Field, 2013: 442–444). The significant differences in the means between the groups are reported in Table 5.

Table 5.

Results of ANOVA, political leaders. Rate the following dimensions on a scale 1–7

Scale
1 unanimous decision – 7 vote	F(2, 157) = 6.72, MSE = 12.98, p = .002
1 based on information – 7 based on emotions	F(2, 158) = 11.44, MSE = 19.02, p = .000
1 public administrator-led – 7 political leader-led	F(2, 158) = 9.74, MSE = 16.10, p = .000
1 trust – 7 distrust	F(2, 156) = 28.46, MSE = 42.74, p = .000
1 bureaucratic – 7 flexible	F(2, 158) = 13.56, MSE = 24.71, p = .000

The trustful group (M = 3.58) viewed the decisions to be made more often unanimously than did the critical group (M = 4.75). The trustful group (M = 3.13) found the decision-making to be more based on information than did the two other groups do (middle-of-the-road M = 4.20, critical M = 4.26). The critical group (M = 2.06) found the decision-making process to be more public administrator-led than did the two other groups (middle-of-the-road M = 3.07, trustful M = 3.30). When considering the scale trust – distrust, significant differences emerged between all three groups (trustful M = 3.13, middle-of-the-road M = 4.06, critical M = 5.29). Likewise, considering the scale bureaucratic – flexible significant differences emerged between all three groups (critical M = 2.47, middle-of-the-road M = 3.42, trustful M = 4.09). The critical group members pictured the decision-making as most bureaucratic. For the critical group, the overall picture of decision-making processes is very different from the trustful group. There is more distrust between the actors, more voting and emotions and it is considered to be more bureaucratic and public administrator-led.

For the two groups of public administrators none of the tests were significant. On the dimensions 1 unanimous decision – 7 vote (M = 3.28) and 1 public administrator-led – 7 political leader-led (M = 3.61) there were no differences in the means between the groups of presenting public administrators and non-presenting public administrators. The greatest difference in means was on the 1 bureaucratic – 7 flexible scale (presenting M = 4.13, non-presenting M = 2.92), so the presenting public administrators pictured the decision-making to be slightly more flexible than did the other public administrators. On the dimensions 1 based on information – 7 based on emotions (presenting M = 3.74, non-presenting M = 4.31) and 1 trust – 7 distrust (presenting M = 3.30, non-presenting M = 3.92) the differences between the groups were smaller. However, the presenting public administrators viewed the decision making to be based slightly more on trust and information than the other public administrators did.

Conclusions and discussion

In today’s world, the change in public organisations is constant, sometimes even very sudden as, for example, the effects of the global economy are mirrored on the local level. The interplay of structure and agency is constantly present in the relationship and we are interested in the ways people act in the decision-making processes. The possibilities offered by the research material of this study have their limits. The strong contextual element and small data sets can be considered to be a limitation to this study. However, the study is experimental in nature and we argue that it can still reveal interesting phenomena in the field of interaction between political leaders and public administrators and provide interested scholars with ideas for new research designs. With this study design and these data sets it is not feasible to compare the two groups of actors, but the views of both groups are still valuable for the overall picture on interaction.

The data were collected from six local government organisations, which have experienced a turbulent operating environment and extensive organisational reforms during at least five years before the data collection. Thus, the case organisations represent well the reality on the Finnish municipal sector (Kettunen, 2015). The rapid contextual change requires prompt and well-informed action from both political leaders and public administrators; even political differences can be pushed aside when needed. As the reforms and changes in the local government operating environment are a commonplace, paradigmatic change in local government strategic management and the relationship between political leaders and public administrators can be expected.

First, we asked if the political leaders and public administrators have different kinds of interaction patterns and, if so, what kind of patterns are they. We discerned three different kinds of interaction patterns for political leaders and formed three groups (trustful, middle-of-the-road and critical) based on five factors (lack of trust, municipal residents, prejudice, information process, strategic soundness) measuring the dimensions of decision-making processes. There were no differences between the groups on political affiliations, but there were slight differences in the domiciles and employment status. Personal qualities and skills remain important to interaction (see also e.g. Demir, 2009, 2011; Demir and Nank, 2012; Jacobsen, 2006), the three groups we formed seem to represent different ways of interacting in the decision-making process. For public administrators, we considered the institutional position as presenting or non-presenting public administrator to be a possible grouping factor and thus, wanted to see if that would be the case when exploring the empirical data.

Considering the second research question on what effect the possible patterns have on the ways in which the political leaders and public administrators work, we observed that there seems to be differences in the way the different groups of political leaders work and cooperate with other actors in the decision-making process. For example, there was a large difference in the opinions between critical and trustful groups on personal relations having an effect on decision-making, where the critical group mostly agreed and the trustful group mostly disagreed. The critical group also pictured the decision-making to be based more on emotions than the trustful group did. It might indicate that, whereas the formal structure surely is important (see e.g. Gulick, 1937; Simon, 1957) when considering the interaction between political leaders and public administrators, the personal patterns of interaction do play a role in how the political leaders act. This interpretation strengthens the idea of interplay of structures and action. There is also some previous evidence on the structures having an effect on the interaction patterns between political leaders and public administrators. The traditionally strong ties between the committees and public administrators in the single service departments weakened and the relations between the political and administrative leadership strengthened when a new executive board approach to the organisation was adopted in Denmark (Bækgaard, 2011). This study shows that decision-making processes in local government are not only dependent on the political views of the actors, but that the actors also have different interaction patterns, which may have an effect on the way they work in the turbulent setting. Even if one of these groups was labelled critical, it does not mean that the contribution of the group in the decision-making is necessarily negative as the strategic management and decision-making processes need contrasts to function.

The institutional position public administrators hold seems to have an effect on how they consider the decision-making processes to function. Most presenting public administrators disagreed with the statement that personal relations have an effect on decision-making. This is possibly an indication of functioning and good personal relations, which may lead to more interaction and information sharing. The seminars, which both political leaders and public administrators considered as important, are an interesting tool in decision-making. They are institutionalised, but the nature of communication in the seminars is freer than in the official committee, board or council meetings.

When designing the study, we assumed that the possible individual interaction patterns which both political leaders and public administrators have may have an effect on the ways they view the local government decision-making processes. This was measured with five variables assessing the picture on bipolar descriptions. Even if the scales are similar for both groups, a comparison is not feasible as the groups vary so much in size (N = 166 and N = 37). However, we consider it to be important to explore both sides of the local government decision-making system. As Peters and Pierre (2003) put it, politics and administration are different elements of the same process. Political leaders and public administrators are interdependent. However, the differing basis of their positions is visible in their work and views on the decision-making processes. Above that, the individual interaction patterns discerned in this study have an effect on how the actors picture the decision-making process.

Considering the political leaders, there were statistically significant differences between the groups of critical, middle-of-the-road and trustful representatives. In general, the trustful group considered the decision-making processes as information-based, trustworthy, unanimous and flexible whereas the critical group considered the processes to include more emotion-based elements, distrust and voting combined with more bureaucratic and public administrator-led working methods.

The groups of public administrators were very small, so no statistically significant differences existed in the data. However, there were slight differences between the groups and the presenting public administrators pictured the decision-making to be based slightly more on trust and information than did the other public administrators. A possible explanation for this might be that presenting public administrators are in close contact with political leaders because of their duties.

Local government is especially in the Finnish context very close to the municipal resident as it has the responsibility of organising majority of the social, health and educational services. The political leaders and public administrators in local government face a dilemma; how to respond to the external change and manage the internal reforms still being reliable and responsible from the perspective of the municipal residents? For this kind of response, both formal and informal communication between the two groups is needed. The tendency revealed in an experimental study can later be tested with other data sets to refine the idea, even a comparative study between the two groups of actors could be feasible with a different research design.

We believe that the interaction mechanisms discovered in this study are not only valid in the Finnish context, but also interesting and exist in other contexts. The relationship between political leaders and public administrators is a multi-level one. Power, responsibility, ethical consideration, flow of information and personal relations are at the centre of the local government decision-making processes. Understanding the relationship between reform and interaction in decision-making helps not only researchers but also people in practical management and decision-making positions.

Footnotes

Acknowledgements

We thank the Foundation for Municipal Development (Kunnallisalan kehittämissäätiö) for funding the research project ‘Dualistic Management: Interaction Between Political Leaders and Public Administrators in Local Government’.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Archer

(1995) Realist Social Theory: The Morphogenetic Approach, Cambridge: Cambridge University Press.

Askim

(2007) How do politicians use performance information? An analysis of the Norwegian local government experience. International Review of Administrative Sciences 73(3): 453–472.

Bækgaard

(2011) The impact of formal organizational structure on politico-administrative interaction: Evidence from a natural experiment. Public Administration 89(3): 1063–1080.

Baruch

Holtom

(2008) Survey response rate levels and trends in organizational research. Human Relations 61(8): 1139–1160.

Berg R and Rao N (eds) (2005) Transforming Local Political Leadership. London: Palgrave MacMillan.

Bergström

Magnusson

Ramberg

(2008) Through a glass darkly: Leadership complexity in Swedish Local Government. Local Government Studies 34(2): 203–220.

Bryant

Yarnold

(1995) Principal-components analysis and confirmatory factor analysis. In: Grimm

Yarnold

(eds) Reading and Understanding Multivariate Statistics, Washington, DC: American, pp. 99–136.

Christensen

Lægreid

(2002) Complex Patterns of Interaction and Influence Among Political and Administrative Leaders, Bergen: Stein Rokkan Centre for Social Studies, Bergen University Research Foundation.

Constitution of Finland 731/1999.

10.

Copus C (2012) The role of local governance in Europe. In: Wilhelm Hofmeister W, Sarmah M and Rüppe P (eds) Local Politics and Governance, Panorama: Insights into Asian and European Affair. Singapore: Konrad Adenauer Stiftung, pp.39–48.

11.

Dalton

(2008) Citizenship norms and the expansion of political participation. Political Studies 56(1): 76–98.

12.

Demir

(2009) The complementarity view: Exploring a continuum in political–administrative relations. Public Administration Review 69(5): 876–888.

13.

Demir

(2011) Interaction frequency and quality as two dimensions of complementarity: An empirical examination of some contingency variables. Public Organization Review 11(3): 265–281.

14.

Demir

Nank

(2012) Interaction quality in political–administrative relations in the United States: Testing a multi-dimensional model. International Journal of Public Administration 35(5): 329–339.

15.

Demir

Reddick

(2012) Understanding shared roles in policy and administration: An empirical study of council–manager relations. Public Administration Review 72(4): 526–536.

16.

Denters B and Rose L (2005) Local governance in the third millennium: A brave new world? In: Denters B and Rose L (eds) Comparing Local Governance: Trends and Developments. London: Palgrave MacMillan.

17.

Fabrigar

Wegener

MacCallum

et al. (1999) Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods 4(3): 272–299.

18.

Field

(2013) Discovering Statistics Using IBM SPSS Statistics, 4th ed. London: Sage.

19.

Follett

(1973) The process of control. In: Gulick

Urwick

(eds) Papers on the Science of Administration, New York: Columbia University, Institute of Public Administration.

20.

Giauque

Resenterra

Siggen

(2009) Modernization routes and relations between political and administrative spheres in Switzerland. International Review of Administrative Sciences 75(4): 687–713.

21.

Guba

Lincoln

(1994) Competing paradigms in qualitative research. In: Denzin

Lincoln

(eds) Handbook of Qualitative Research, London: Sage, pp. 105–117.

22.

Gulick

(1937) Notes to the theory of organizations. With specific reference to government. In: Gulick

Urwick

(eds) Papers on the Science of Administration, New York: Columbia University, Institute of Public Administration.

23.

Hair

Jr Black

Babin

et al. (2014) Multivariate Data Analysis: Pearson New International Edition, Harlow: Pearson Education Limited.

24.

Hansen

Niels

(2002) The relationship between politicians and administrators – A logic of disharmony. Public Administration 80(4): 733–750.

25.

Harisalo

(2008) Organisaatioteoriat, Tampere: Tampere University Press.

26.

Jacobsen

(2005) Sand in the machinery? Comparing bureaucrats’ and politicians’ attitudes toward public sector reform. European Journal of Political Research 44(6): 767–799.

27.

Jacobsen

(2006) The relationship between politics and administration: The importance of contingency factors, formal structure, demography, and time. Governance: An International Journal of Policy, Administration, and Institutions 19(2): 303–323.

28.

Horvath A and Paolini G (2013) Political participation and EU citizenship: Perceptions and behaviours of young people. Evidence from Eurobarometer surveys. Report produced by the Education, Audiovisual and Culture Executive Agency (EACEA).

29.

Kerlinger

(1986) Foundations of Behavioral Research, New York: Holt, Rinehart and Winston.

30.

Keski-Petäjä

(2009) Merkintöjä kunnan organisoitumisesta. In: Haatainen

Keski-Petäjä

and Vartola

(eds) Muuttuva hallinto ei sammaloidu, Tampere: Juvenes Print – Tampereen yliopistopaino Oy, pp. 65–92.

31.

Kettunen

(2015) The Finnish Municipal Reform. Annals of the Croatian Political Science Association: Political Science Journal 11(1): 55–70.

32.

Kurikka P (2014) Municipal Management System. Helsinki: The Association of Finnish Local and Regional Authorities. Available at: www.localfinland.fi/en/authorities/local-self-government/Managementsystem/Pages/default.aspx.

33.

Lægreid

Roness

(1999) Administrative reform as organized attention. In: Egeberg

Lægreid

(eds) Organizing Political Institutions, Oslo: Scandinavian University Press.

34.

Local Government Act 519/2007.

35.

Long NE (1949) Power and administration. Public Administration Review 9(4)(Autumn): 257–264.

36.

McAuley

Duberley

Johnson

(2007) Organization Theory. Challenges and Perspectives, Harlow: Pearson.

37.

March

(1981) Decisions in organizations and theories of choice. In: Van de Ven

Joyce

(eds) Perspectives on Organization Design and Behavior, New York: Wiley Interscience.

38.

Mason

Mitroff

(1981) Challenging Strategic Planning Assumptions, New York: John Wiley & Sons.

39.

Mintzberg

Raisinghani

Theoret

(1976) The structure of unstructured decision processes. Administrative Science Quarterly 21: 246–276.

40.

Molina-Azorin

(2012) Mixed methods research in strategic management: Impact and applications. Organizational Research Methods 15(1): 33–56.

41.

Mouritzen PE and Svara J (2002) Leadership at the Apex. Politicians and Administrators in Western Local Governments. Pittsburgh: University of Pittsburgh Press. Available at: http://digital.library.pitt.edu/cgi-bin/t/text/text-idx?c=pittpress;cc=pittpress;view=toc;idno=31735046143545.

42.

Niiranen

Joensuu

(2014) Political leaders and public administrators: Key values and stumbling blocks in decision-making and interaction. In: Joyce

Drumaux

(eds) Strategic Management in Public Organizations, European Practices and Perspectives, New York: Routledge.

43.

Niiranen

Joensuu

Martikainen

(2013) Millä tiedolla kuntia johdetaan?, Sastamala: Vammalan kirjapaino.

44.

Olsen

(1983) Organized Democracy, Bergen: Universitetsförlaget.

45.

Peters

Pierre

(2003) Introduction: The role of public administration in governing. In: Peters

Pierre

(eds) Handbook of Public Administration, London: Sage.

46.

Pollitt

Bouckaert

(2004) Public management reform: A comparative analysis, 2nd ed. Oxford: Oxford University Press.

47.

Pollitt C, Van Thiel S and Homburg V (eds) (2007) New Public Management in Europe: Adaptations and Alternatives. Basingstoke: Palgrave Macmillan.

48.

Quinn

(1980) Strategies for Change: Logical Incrementalism, Homewood, IL: Richard D. Irwin.

49.

Rapeli

(2010) Tietääkö kansa? Kansalaisten politiikkatietämys teoreettisessa ja empiirisessä tarkastelussa. Annales Universitatis Turkuensis. Sarja C, Osa 295, Turku: Turun yliopisto.

50.

Ring

Perry

(1985) Strategic management in public and private organizations: Implications of distinctive contexts and constraints. Academy of Management Review 10(2): 276–286.

51.

Selznick

(1957) Leadership in Administration, Berkeley: University of California Press.

52.

Shrivastava

Nachman

(1989) Strategic leadership patterns. Strategic Management Journal 10: 51–66.

53.

Simon

(1957) Administrative Behaviour, New York: Macmillan.

54.

Simon

Smithburg

Thompson

(1950) Public Administration, New York: Alfred Knopf.

55.

Sloam

(2013) Voice and equality: Young people's politics in the European Union. West European Politics 36(3): 1–23.

56.

State Treasury (2015) Available at: www.Suomi.fi/state_and_municipalities (accessed 21 October 2015).

57.

Strandman

(2009) Se vain ilmestyi – Vuorovaikutukseen perustuva strategian viestintä kuntaorganisaatiossa. Acta Universitatis Lapponiensis 168, Rovaniemi: Lapin yliopisto.

58.

Svara

(1998) The politics-administration dichotomy model as aberration. Public Administration Review 58(1): 51–58.

59.

Svara

(1999) Complementarity of politics and administration as a legitimate alternative to the dichotomy model. Administration & Society 30: 676–705.

60.

Szücs

Strömberg

(2009) The more things change, the more they stay the same: The Swedish Local Government Elite between 1985 and 2005. Local Government Studies 35(2): 251–270.

61.

Tabachnick BG and Fidell LS (2007) Using Multivariate Statistics. 5th edn. Boston: Pearson/Allyn&Bacon.

62.

Verheul

Schaap

(2010) Strong leaders? The challenges and pitfalls in mayoral leadership. Public Administration 88(2): 439–454.

63.

Walgenbach

Meyer

(2008) Neoinstitutionalistische Organisationstheorie, Stuttgart: W. Kohlhammer.

64.

Weber

(1922) Bureaucracy. In: Shafritz

Hyde

(eds) Classics of Public Administration (2008), 6th ed. United States: Wadsworth Publishing Company, pp. 50–55.

65.

Wilson

(1887) The study of administration. Political Science Quarterly 2: 209–213.

66.

Wynn

Jr Williams

(2012) Principles for conducting critical realist case study research in information systems. MIS Quarterly 36(3): 787–810.